Skip to content

Emailchaser MCP server

com.emailchaser/emailchaseremailchaser.comstreamable-httpv1.0.1Scanned

Run cold email outbound from an AI agent: campaigns, leads, replies, sender accounts, autopilot.

102 tools: 21 can delete or cancel things, 1 do something that can't be undone, 33 change things and 47 only read.

Tools: 21 delete, 1 irreversible, 33 write, 47 read.
  • Reads private data · 29
  • Reads text others write · 46
  • Can send data out · 12
  • 5 findings

Tools listed without sign-in. Remote posture

With Anlyon

21 tools here can delete or cancel things and 1 does something that can't be undone. Govern them.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the Emailchaser MCP server can do

The Emailchaser MCP server is published by emailchaser.com, a domain the MCP Registry verified. It has 102 tools. It can delete or cancel things (21 tools: launch_campaign, replace_campaign_sequence, delete_campaign, detach_campaign_sender_email and 17 more), do things that cannot be undone, such as sending, paying or publishing (update_reply_draft) and create or change data (33 tools: confirm_connection, create_campaign, update_campaign, pause_campaign and 29 more). 47 tools only read.

29 tools read private data (list_campaigns, get_campaign, get_campaign_sequence and 26 more); 46 tools bring in text other people can write (list_campaigns, get_campaign, launch_campaign and 43 more); 12 tools can send data out of your system (attach_campaign_sender_emails, detach_campaign_sender_email, update_lead and 9 more). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.

The fixed checks matched 5 findings (long-description, annotation-conflict and tool-shadowing). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in, and its tools changed 2 times since Anlyon Radar first saw them.

Last scanned Nov 13, 2026, version 1.0.1.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

102 tools

Deletes or cancels 21Claude Code: deny
  • add_lead_finder_prospectsSPENDS CREDITS, AND BY DEFAULT METERED EMAIL VERIFICATION.Reads text others write1 finding

    Why it is in this group

    • The name "add_lead_finder_prospects" says "add": a word for a change.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Info

      The description is 2,681 characters long, over the 2,000 this check looks for.

      Matched text

      description ends: "...an importId, and the add's import record keeps changing until finishedAt is set."

      What this check looks for
    Description, as the server wrote it
    SPENDS CREDITS, AND BY DEFAULT METERED EMAIL VERIFICATION. Adds people from the Lead Finder contact database to a campaign as leads, revealing their contact details, in the background. Pass refs (from Lead Finder search results; up to 1,000 different people by default) to add exactly those people, or pass filters and count (1 to 5,000 by default) to add count people matching the filters who are not in the workspace yet. A filters add always walks the matches from the top and skips anyone already a lead, free and not counted, so repeating the same add adds the NEXT count people and charges again. Never repeat an add to retry: an add that errors or times out may still have started, and adding again can add the same people twice. A filters add stops early when the audience runs out or once it has looked at five people for every one requested (at least 1,000). People already added count toward that limit even though skipping them is free, so after an add of more than about 1,000 people a small follow-up can stop with nobody added. Every person a filters add looks at, added or skipped, also uses the account's daily allowance for filters adds (25,000 people by default). Each person added costs 1 credit ($0.033 at list), reported as creditsPerProspect; the response reports estimatedCredits, the most the add can cost in credits, and the wallet must hold that much for it to start. People already in the workspace, blocklisted, without a usable address, on a personal mailbox when the campaign only takes business addresses, or marked invalid by verification are skipped and cost no credits. verifyEmails (on by default) checks every screened address with the paid email verification waterfall, including the ones it marks invalid: one or two checks per address, billed as metered usage on the next invoice (price per check: GET /email-verification/rates in the REST API), within the account's monthly verification spend limit. Catch-all and unconfirmed addresses are added and charged, and so is every address once that limit is reached; verification.limitReached and verification.unknown on the add's import record count them. Leads added to a running or paused campaign get their emails at once (a paused campaign sends them when resumed); adding to a completed campaign resumes it, so it starts sending again; in a draft campaign the emails are created at launch. At most 3 adds run at once in a workspace. A refusal names its reason, and a 429 also says when to retry (reason running_imports, daily_import_limit, daily_budget, monthly_budget or fair_use_floor). The response carries an importId, and the add's import record keeps changing until finishedAt is set.

    Annotations the server declares

    • title: Lead Finder: add people to a campaign (spends credits)
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • approve_autopilot_runApproves an autopilot run that is awaiting approval — the one human gate in the flow.Reads text others writeNo findings

    Why it is in this group

    • The name "approve_autopilot_run" says "approve": a word for a change.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Approves an autopilot run that is awaiting approval — the one human gate in the flow. CAN SPEND REAL MONEY: approval first reveals the run's prospects at 1 credit ($0.033 at list) each, then, if the workspace has no connected sending account, places one done-for-you order sized from the plan snapshotted at start, at most 10 .com domains and at most six mailboxes per domain, never more mailboxes than the plan, charged at the prices in effect when the order is placed. So the plan's mailbox and domain counts are a ceiling and its dollar figures an estimate, not an exact quote. No credits are bought, but the reveal and later top-ups spend the wallet's credits. If a sending account is already connected, no infrastructure is bought; a run started without budgetUsd and with no connected account fails. Show the user the plan returned when the run was started before calling this. Nothing is revealed, bought or sent before this call; after it, sending starts without further confirmation.

    Annotations the server declares

    • title: Autopilot: approve a run
    • readOnlyHint: false
    • destructiveHint: true
    • openWorldHint: true
  • cancel_email_verification_jobStops a running verification job.1 finding

    Why it is in this group

    • The name "cancel_email_verification_job" says "cancel": a word for removing or ending something.
    • The server marks it not destructive (destructiveHint: false), but an annotation never lowers a class.

    Findings

    • Warn

      The server marks "cancel_email_verification_job" not destructive (destructiveHint: false), and its own name, description or input give the class "destructive".

      Matched text

      annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "cancel_email_verification_job" says "cancel": a word for removing or ending something.

      What this check looks for
    Description, as the server wrote it
    Stops a running verification job. Addresses already verified stay in the results and stay billed; addresses that never ran are never charged, so cancelling costs nothing further.

    Annotations the server declares

    • title: Cancel an email verification job
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • cancel_lead_finder_importStops a running Lead Finder add.No findings

    Why it is in this group

    • The name "cancel_lead_finder_import" says "cancel": a word for removing or ending something.
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Stops a running Lead Finder add. It reads canceled at once. A batch the worker is already writing is still added and charged; a batch still being screened or verified is dropped, though verification checks already made are billed. Credits held for the rest are released. Until finishedAt is set, added, creditsSpent and even status can still change: if that batch was the add's last, it ends completed or failed instead. People already added stay in the campaign. A stopped add cannot be resumed: adding the people again starts a new add. Stopping one that already finished changes nothing.

    Annotations the server declares

    • title: Lead Finder: stop an add
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • create_dfy_orderSPENDS REAL MONEY.Reads text others writeNo findings

    Why it is in this group

    • The name "create_dfy_order" says "create": a word for a change.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    SPENDS REAL MONEY. Places a real order for sending domains and pre-warmed mailboxes: domains are registered at their listed price (about $13.99/year for a .com; each domain has its own listed price) and each mailbox costs about $3 setup plus $6/month. The order is accepted immediately and provisioned in the background, so the order's status changes over time. Purchased domains redirect visitors to forwardingDomain. Confirm the domains and their prices with the user before ordering. Every domain in domains needs at least one mailbox in mailboxes whose domainName is that domain, or the order is refused before anything is bought; mailboxes can also go on a domain from an earlier order.

    Annotations the server declares

    • title: Create done-for-you order (charges real money)
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • create_email_verification_jobSPENDS MONEY: every address checked adds metered usage to the workspace's next invoice.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "create_email_verification_job" says "create": a word for a change.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe name "create_email_verification_job" says "create" and "email": it writes something other people read.
    Description, as the server wrote it
    SPENDS MONEY: every address checked adds metered usage to the workspace's next invoice. This does NOT use prospect credits and it does NOT create or send a campaign — it verifies a list on its own and gives back a per-address verdict you can download as CSV. An address costs one verification credit when the first provider rejects it outright and two when both providers answer, so budget two credits per address in the list. An address that gets no verdict is NOT billed and comes back with result unknown. Duplicates are removed and unparseable entries are returned in invalid_emails before anything is charged, so the total in the response is what will be billed against, not what you submitted. Needs an active or trialing subscription: a past_due workspace is refused with subscription_not_active, and a trialing workspace may submit up to 100 addresses in total across all its jobs, after which it is refused with trial_allowance_exceeded (the message says how many are left). If the workspace's monthly verification allowance runs out mid-job the job does not fail — the remaining addresses come back unknown, marked, still in the download, and unbilled. The job runs in the background: its state reaches done when every address has a verdict, and the verdicts are then available per address.

    Annotations the server declares

    • title: Verify a list of email addresses (spends money)
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • delete_campaignPermanently deletes a campaign and all associated data, including its emails, sequences and scheduled tasks.No findings

    Why it is in this group

    • The name "delete_campaign" says "delete": a word for removing or ending something.
    • The description says "Permanently deletes".
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Permanently deletes a campaign and all associated data, including its emails, sequences and scheduled tasks. This cannot be undone.

    Annotations the server declares

    • title: Delete campaign
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • delete_icpDeletes an Ideal Customer Profile.No findings

    Why it is in this group

    • The name "delete_icp" says "delete": a word for removing or ending something.
    • The description says "cannot be undone".
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Deletes an Ideal Customer Profile. Deleting the primary leaves the workspace without an active profile. This cannot be undone.

    Annotations the server declares

    • title: Delete ICP
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • delete_leadPermanently deletes a lead by ID.No findings

    Why it is in this group

    • The name "delete_lead" says "delete": a word for removing or ending something.
    • The description says "Permanently deletes".
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Permanently deletes a lead by ID. This cannot be undone.

    Annotations the server declares

    • title: Delete lead
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • delete_webhookDeletes a registered webhook endpoint by ID so it stops receiving events.No findings

    Why it is in this group

    • The name "delete_webhook" says "delete": a word for removing or ending something.
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Deletes a registered webhook endpoint by ID so it stops receiving events.

    Annotations the server declares

    • title: Delete webhook
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • detach_campaign_sender_emailDetaches a sender email from a campaign so the campaign stops sending through that mailbox.Can send data outNo findings

    Why it is in this group

    • The name "detach_campaign_sender_email" says "detach": a word for a change.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • outboundThe name "detach_campaign_sender_email" says "detach" and "email": it writes something other people read.
    Description, as the server wrote it
    Detaches a sender email from a campaign so the campaign stops sending through that mailbox. Unsent follow-ups to emails this sender already sent are canceled and do not come back on re-attach; a running campaign is rescheduled onto its remaining senders; already-sent emails are untouched. Idempotent: detaching a sender that is not attached is a no-op.

    Annotations the server declares

    • title: Detach sender email from campaign
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • kill_autopilot_runEngages the kill switch: the run halts wherever it is and can NEVER be resumed, and any campaign the run launched is paused so nothing more sends.No findings

    Why it is in this group

    • The name "kill_autopilot_run" says "kill": a word for removing or ending something.
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Engages the kill switch: the run halts wherever it is and can NEVER be resumed, and any campaign the run launched is paused so nothing more sends. A paused run can be resumed later; a killed one cannot. The optional reason is recorded in the run's audit trail.

    Annotations the server declares

    • title: Autopilot: kill a run
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • launch_campaignLaunches a new campaign or resumes a paused one.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Launches a new campaign or resumes a paused one. The campaign configuration is validated before launching; emails start sending per the campaign schedule.

    Annotations the server declares

    • title: Launch campaign
    • readOnlyHint: false
    • destructiveHint: true
    • openWorldHint: true
  • purchase_creditsSPENDS REAL MONEY.Reads text others writeNo findings

    Why it is in this group

    • The name "purchase_credits" says "purchase": a word that usually means money moves, which cannot be taken back.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    SPENDS REAL MONEY. Immediately charges the workspace's saved default card (the one behind the active subscription), off-session, with no confirmation step beyond this call — every successful call is a new charge. Buys 1,000 to 10,000 prospect credits at the tiered list price: $33/1k · $20/1k from 5k. The server prices the charge, so treat this as the list price and not a quote. Errors carry a machine-readable code: billing_required and payment_failed mean no money moved (fix billing or the card, then retry); credits_free means this workspace's credits are free, so nothing was charged and there is nothing to buy; temporarily_unavailable means the purchase stopped before any charge, so it is safe to retry shortly; purchase_incomplete means the card WAS charged but crediting failed — do NOT retry, support is already notified; purchase_unconfirmed means the outcome is unknown or the purchase completed without a balance to report: a stripe_topup entry in the credit ledger means the card was charged, so retrying before the outcome is known can charge twice. After a timeout or a dropped connection the outcome is equally unknown, so do not call this again until it is known.

    Annotations the server declares

    • title: Buy credits (charges real money)
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • remove_blocklist_entriesUnblocks up to 1,000 domains or email addresses in one call, so campaigns can email them again.No findings

    Why it is in this group

    • The name "remove_blocklist_entries" says "remove": a word for removing or ending something.
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Unblocks up to 1,000 domains or email addresses in one call, so campaigns can email them again. Pass values (domains or addresses, matched case-insensitively) and/or ids (blocklist entry IDs); a value that is not blocked is counted as skipped rather than failing the call. Scope defaults to "workspace" and never touches the account-wide list unless asked; removing an account-wide entry only works from the main workspace.

    Annotations the server declares

    • title: Remove blocklist entries
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • remove_blocklist_entryRemoves one suppression entry by ID, so campaigns can email that domain or address again.No findings

    Why it is in this group

    • The name "remove_blocklist_entry" says "remove": a word for removing or ending something.
    • The description says "cannot be undone".
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Removes one suppression entry by ID, so campaigns can email that domain or address again. This cannot be undone: the entry and the record of when it was added are gone. A sub-workspace key cannot remove an account-wide entry it merely inherits. It removes one entry per call.

    Annotations the server declares

    • title: Remove one blocklist entry
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • replace_campaign_sequenceReplaces a campaign's WHOLE sequence with the steps provided.No findings

    Why it is in this group

    • The name "replace_campaign_sequence" says "replace": a word for a change.
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Replaces a campaign's WHOLE sequence with the steps provided. This is not a partial edit: steps you leave out are removed, so send the full sequence every time. Sending the same payload twice leaves the same result, so a retry cannot duplicate steps. Editing a running campaign is allowed and changes future sends only; already-sent emails are untouched.

    Annotations the server declares

    • title: Replace campaign sequence
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
  • send_replyAnswers an inbound reply with your own message, in the same email thread and from the mailbox the conversation runs through (in a conversation the prospect started, your mailbox th…Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "send_reply" says "send": a word for sending, and what is sent cannot be taken back.
    • The input addresses people (input.to): what it sends cannot be taken back.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe name "send_reply" says "send": it sends or publishes something to other people.
    Description, as the server wrote it
    Answers an inbound reply with your own message, in the same email thread and from the mailbox the conversation runs through (in a conversation the prospect started, your mailbox they wrote to). It works for every inbound reply, including the ones Emailchaser writes no AI draft for (wrong_person, uncategorized), and it is how you send a second message in a conversation that went quiet: answer the prospect's latest reply again. The message is scheduled at once, goes out within about a minute and cannot be recalled; its status then moves from scheduled to sent in the lead's conversation. Only id and body are required. Leave to out to Reply All: the person who wrote goes in To and everyone else they addressed in Cc, never one of your own mailboxes. Set to to choose the recipients yourself: then to, cc and bcc are used exactly as given (cc is only read together with to; bcc works either way), at most 20 addresses in all. There is no subject: the answer keeps the thread's subject with "Re: " in front, so the thread never breaks. contentType is "text" by default; use "html" only when body is HTML (paragraphs, line breaks, bold, italics, links, lists and headings are kept; images, tables and inline styles are removed). Merge tags such as {first_name} and spintax such as {Hi|Hello} are filled in when it sends. No signature is added: put yours in body. attachments takes at most 5 files, 10 MB in total. Refused with 409 when the id is not an inbound reply or an identical body went to it less than 60 seconds ago (so a retry after a timeout never sends twice), 422 when there is no connected sender mailbox or recipient, the lead has unsubscribed or an earlier email to the recipient hard-bounced, and 429 after 20 answers to one reply in an hour. Needs a read & write key. API reference: https://www.emailchaser.com/docs/api-reference/replies

    Annotations the server declares

    • title: Send your own reply
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • send_reply_draftSends an AI reply draft to the prospect through the conversation's sender mailbox; it is scheduled for delivery right away and cannot be recalled.Reads text others writeCan send data outNo findings

    Why it is in this group

    • The name "send_reply_draft" says "send": a word for sending, and what is sent cannot be taken back.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe name "send_reply_draft" says "send": it sends or publishes something to other people.
    Description, as the server wrote it
    Sends an AI reply draft to the prospect through the conversation's sender mailbox; it is scheduled for delivery right away and cannot be recalled. The draft's CURRENT subject and body are what goes out, so check its wording before sending: an edit has to be made before it is sent. Refused with 409 for anything that is not a sendable AI reply draft (already sent or scheduled emails, inbound replies, sequence templates) and with 422 when the conversation has no connected sender mailbox or no resolvable recipient.

    Annotations the server declares

    • title: Send AI reply draft
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • source_prospectsSPENDS CREDITS.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it destructive (destructiveHint: true).

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    SPENDS CREDITS. Queues a background job that searches Emailchaser's contact database with an Ideal Customer Profile's targeting criteria, reveals the matching people and adds them to the campaign as leads. Uses the workspace's primary ICP when icpId is omitted, and adds 50 prospects unless count says otherwise (max 500 per call). Every stored prospect costs the reveal price in credits (1 at the time of writing; the response reports creditsPerProspect and estimatedCredits, the most this batch can cost). Duplicates, blocklisted domains and contacts without an email address are filtered out before any credit is spent. Asynchronous: the prospects arrive in the campaign's leads over time. Calling again for the same campaign and profile pages deeper into the audience instead of re-revealing (and re-paying for) the same people.

    Annotations the server declares

    • title: Source prospects into a campaign (spends credits)
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: false
    • openWorldHint: true
  • unmark_lead_meetingRemoves the booked-meeting mark from a lead, e.g.No findings

    Why it is in this group

    • The description's first verb is "remove": a word for removing or ending something.
    • The server marks it destructive (destructiveHint: true).
    Description, as the server wrote it
    Removes the booked-meeting mark from a lead, e.g. when it was recorded by mistake or the meeting was canceled: clears meetingBookedAt and, when the category is still meeting_booked, reverts it to interested. The original booked time is lost. Idempotent: unmarking a lead with no meeting is a no-op.

    Annotations the server declares

    • title: Unmark booked meeting on lead
    • readOnlyHint: false
    • destructiveHint: true
    • idempotentHint: true
    • openWorldHint: false
Can't be undone 1Claude Code: ask
  • update_reply_draftEdits the subject and/or body of an AI reply draft before it goes out.Can send data outNo findings

    Why it is in this group

    • The name "update_reply_draft" says "update": a word for a change.
    • The name "update_reply_draft" also says "reply": a word for sending, and what is sent cannot be taken back.

    What it can reach

    • outboundThe name "update_reply_draft" says "reply": it sends or publishes something to other people.
    Description, as the server wrote it
    Edits the subject and/or body of an AI reply draft before it goes out. Provide at least one field; provided fields must be non-empty. This never changes the draft's status or sends anything: the draft stays a draft until it is sent. Emails that are not AI reply drafts (sent or scheduled emails, inbound replies, sequence templates) are refused. API reference: https://www.emailchaser.com/docs/api-reference/replies

    Annotations the server declares

    • title: Edit AI reply draft
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
Changes things 33Claude Code: ask
  • add_blocklist_entriesAdds up to 1,000 suppression entries in one call so no campaign ever emails them.No findings

    Why it is in this group

    • The name "add_blocklist_entries" says "add": a word for a change.
    Description, as the server wrote it
    Adds up to 1,000 suppression entries in one call so no campaign ever emails them. Each value is either a full email address (contains @) or a bare domain; the type is inferred per value. Existing entries and invalid values are skipped, so the call is safe to retry and suited to carrying over a suppression list from another sending platform. Scope defaults to "workspace"; "global" blocks across every workspace on the account and only works from the main workspace.

    Annotations the server declares

    • title: Add blocklist entries
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • add_leadsCreates or updates up to 1,000 leads in one request, optionally adding them to a campaign via campaignId.No findings

    Why it is in this group

    • The name "add_leads" says "add": a word for a change.
    Description, as the server wrote it
    Creates or updates up to 1,000 leads in one request, optionally adding them to a campaign via campaignId. Existing leads (matched by email) are updated instead of duplicated. Returns the id of every lead written, so they can be fetched or updated afterwards. Attributes outside the named fields can be passed in customVariables and become merge tags in email copy.

    Annotations the server declares

    • title: Add leads
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • attach_campaign_sender_emailsAttaches connected sender email accounts to a campaign.Can send data outNo findings

    Why it is in this group

    • The name "attach_campaign_sender_emails" says "attach": a word for a change.

    What it can reach

    • outboundThe name "attach_campaign_sender_emails" says "attach" and "email": it writes something other people read.
    Description, as the server wrote it
    Attaches connected sender email accounts to a campaign. A campaign only sends through the senders attached to it, so this call decides which mailboxes it uses. Idempotent: already-attached senders are left untouched and only missing ones are added. The call is refused whole when any requested sender is not a connected sender email in the workspace. Attaching to a running campaign reschedules it so the new senders are picked up.

    Annotations the server declares

    • title: Attach sender emails to campaign
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • confirm_connectionConfirms the API key works and tells Emailchaser which AI assistant connected.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Confirms the API key works and tells Emailchaser which AI assistant connected. Use it when the user connects Emailchaser or asks to confirm or check the Emailchaser connection: it records the assistant and time on the key, which the app's API & MCP page watches to show the user you are in, and completes the 'Connect your AI' onboarding step. Idempotent: repeat calls simply refresh the record. Works with read-only keys as well as read & write keys.

    Annotations the server declares

    • title: Confirm the connection
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • connect_sender_emailConnects an SMTP/IMAP mailbox to the workspace using an app password, with no browser step.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Connects an SMTP/IMAP mailbox to the workspace using an app password, with no browser step. Google and Microsoft mailboxes are NOT supported here because both require an interactive consent screen that cannot be automated; connect those in the app. The address must be on a business domain, so gmail.com and similar are rejected. The returned account's healthScore is null until warm-up has checked 20 of its emails.

    Annotations the server declares

    • title: Connect sender email
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: true
  • copilot_launchCreates a DRAFT campaign from a sequence.Reads text others writeNo findings

    Why it is in this group

    • The description's first verb is "create": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Creates a DRAFT campaign from a sequence. With a LinkedIn Sales Navigator search it also starts finding + verifying leads for it; without one the draft holds the emails and leads can be added from Lead Finder or a CSV upload. Nothing sends — the campaign stays a draft until you review and launch it. Returns the new campaign id, plus a lead-finding job id when lead finding started.

    Annotations the server declares

    • title: Copilot: launch a draft campaign
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
  • copilot_planGiven a company website, the honest AI SDR builds an ideal-customer-profile (ICP) and a suggested cold-email sequence.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Given a company website, the honest AI SDR builds an ideal-customer-profile (ICP) and a suggested cold-email sequence. Read-only — creates nothing. Returns the ICP (personas, titles, industries, suggested Sales Navigator keywords) and a draft sequence you can review.

    Annotations the server declares

    • title: Copilot: plan a campaign
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
  • create_campaignCreates a campaign in the workspace.No findings

    Why it is in this group

    • The name "create_campaign" says "create": a word for a change.
    Description, as the server wrote it
    Creates a campaign in the workspace. The campaign starts as a DRAFT and sends nothing. A new campaign has no emails in it, so it cannot be launched until it has a sequence and a sender email attached. Sending limits and deliverability settings can be set here or changed later.

    Annotations the server declares

    • title: Create campaign
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: false
  • create_icpStores a manually-authored Ideal Customer Profile (ICP): who the workspace sells to, as targeting criteria (titles, seniorities, industries, company sizes, locations, keywords).No findings

    Why it is in this group

    • The name "create_icp" says "create": a word for a change.
    Description, as the server wrote it
    Stores a manually-authored Ideal Customer Profile (ICP): who the workspace sells to, as targeting criteria (titles, seniorities, industries, company sizes, locations, keywords). Set makePrimary to promote it to the workspace's active profile. The profile is stored as given: it is not generated by AI from a website.

    Annotations the server declares

    • title: Create ICP
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: false
  • create_webhookRegisters a webhook endpoint that receives a POST request whenever the selected event happens (email sent, reply received, bounce, lead created, lead category updated, or campaign …Can send data outNo findings

    Why it is in this group

    • The name "create_webhook" says "create": a word for a change.

    What it can reach

    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Registers a webhook endpoint that receives a POST request whenever the selected event happens (email sent, reply received, bounce, lead created, lead category updated, or campaign status changed). API reference: https://www.emailchaser.com/docs/api-reference/webhooks

    Annotations the server declares

    • title: Create webhook
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: false
  • create_workspaceCreates a sub-workspace under the account's main workspace (for example one per client, for an agency) and by default mints a read & write API key bound to it.No findings

    Why it is in this group

    • The name "create_workspace" says "create": a word for a change.
    Description, as the server wrote it
    Creates a sub-workspace under the account's main workspace (for example one per client, for an agency) and by default mints a read & write API key bound to it. The key's fullKey is returned exactly ONCE and cannot be retrieved again, so hand it to the user straight away; it never outlives the calling key. Must be called with the main workspace's key: sub-workspace keys are refused. Workspaces are a Professional-plan feature.

    Annotations the server declares

    • title: Create workspace
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: false
  • create_workspace_api_keyMints a new API key bound to a workspace the calling key already owns: its own workspace, or one of its sub-workspaces when called with the main workspace's key.No findings

    Why it is in this group

    • The name "create_workspace_api_key" says "create": a word for a change.
    Description, as the server wrote it
    Mints a new API key bound to a workspace the calling key already owns: its own workspace, or one of its sub-workspaces when called with the main workspace's key. This is how a workspace whose key was lost gets a new one without a UI step. fullKey is returned exactly ONCE and cannot be retrieved again, so hand it to the user straight away. The new key authorizes the named workspace only and never outlives the calling key. Keys can be revoked in the app under API & MCP.

    Annotations the server declares

    • title: Create API key for a workspace
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: false
  • get_outcomes_reportPuts money against results for a time window: credit spend (grouped by reason, valued at the $0.033 list price) plus done-for-you order costs on one side, and sent emails, replies,…Reads private dataReads text others write1 finding

    Why it is in this group

    • The name "get_outcomes_report" says "get": a word for reading.
    • The description's first verb is "put": a word for a change.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • private-dataThe description's first sentence says "email": it reads mail.
    • untrusted-inputThe description's first sentence says "put" and "email": it reads inbound email, which other people can write.

    Findings

    • Warn

      The server marks "get_outcomes_report" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"openWorldHint":false}; The name "get_outcomes_report" says "get": a word for reading.

      What this check looks for
    Description, as the server wrote it
    Puts money against results for a time window: credit spend (grouped by reason, valued at the $0.033 list price) plus done-for-you order costs on one side, and sent emails, replies, positive replies and booked meetings on the other, with cost per reply, per positive reply and per meeting. Subscription fees are NOT included. campaignId narrows the outcome side only — spend stays workspace-level — so per-campaign cost figures are partial attribution, not a true campaign cost. Use it to answer 'what did we spend and what did we get'.

    Annotations the server declares

    • title: Get outcomes report
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • import_instantlyImports your Instantly.ai campaigns and leads into Emailchaser using your Instantly API key.Reads text others writeNo findings

    Why it is in this group

    • The name "import_instantly" says "import": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Imports your Instantly.ai campaigns and leads into Emailchaser using your Instantly API key. Imported campaigns are created as DRAFTS — nothing sends until you review and launch them. Runs in the background and returns a job id.

    Annotations the server declares

    • title: Import from Instantly
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
  • mark_lead_meetingRecords that a meeting was actually booked with this lead: sets meetingBookedAt to now, sets the lead's category to meeting_booked and fires the LeadCategoryUpdate webhook.No findings

    Why it is in this group

    • The name "mark_lead_meeting" says "mark": a word for a change.
    Description, as the server wrote it
    Records that a meeting was actually booked with this lead: sets meetingBookedAt to now, sets the lead's category to meeting_booked and fires the LeadCategoryUpdate webhook. Emailchaser never infers meetings from reply text or calendars, so this explicit mark is the only way a meeting is counted (in campaign stats totals.meetings and in the outcomes report). Idempotent: repeating the call keeps the original timestamp and fires no second webhook.

    Annotations the server declares

    • title: Mark meeting booked on lead
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • move_lead_to_campaignMoves a lead from one campaign to another: the lead's unsent emails in the source campaign are deleted, and if the target campaign is running, emails for the lead are enqueued ther…No findings

    Why it is in this group

    • The name "move_lead_to_campaign" says "move": a word for a change.
    Description, as the server wrote it
    Moves a lead from one campaign to another: the lead's unsent emails in the source campaign are deleted, and if the target campaign is running, emails for the lead are enqueued there. Use it to re-route a lead into a better-fitting campaign.

    Annotations the server declares

    • title: Move lead to another campaign
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: false
  • pause_autopilot_runSuspends an autopilot run that is still in flight; a paused run can be resumed later.No findings

    Why it is in this group

    • The name "pause_autopilot_run" says "run": a word for a change.
    Description, as the server wrote it
    Suspends an autopilot run that is still in flight; a paused run can be resumed later. Pausing never skips the approval gate. A completed or failed run cannot be paused.

    Annotations the server declares

    • title: Autopilot: pause a run
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: false
  • pause_campaignPauses a running campaign and cancels all currently scheduled emails.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Pauses a running campaign and cancels all currently scheduled emails.

    Annotations the server declares

    • title: Pause campaign
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • plan_autopilotTurns a monthly budget in dollars into a concrete outbound setup: how many domains and mailboxes it buys, monthly sending volume, prospects per month, and costs (one-off setup, rec…No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    Turns a monthly budget in dollars into a concrete outbound setup: how many domains and mailboxes it buys, monthly sending volume, prospects per month, and costs (one-off setup, recurring, first month). creditsUsd, recurringUsd and firstMonthUsd value prospect credits at the list price, 1 credit ($0.033 at list) per revealed prospect, and a note says when that puts the monthly cost above the budget. Pure computation — nothing is created or charged. expectedMeetingsPerMonth is a planning estimate on pessimistic assumptions, not a promise. It answers 'what does $X/month get me' without starting a run.

    Annotations the server declares

    • title: Autopilot: size a budget plan
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: false
  • refresh_icp_audience_sizeRe-counts how many prospects match an ICP's targeting criteria against the live data provider and caches the result on the profile.Reads text others writeNo findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Re-counts how many prospects match an ICP's targeting criteria against the live data provider and caches the result on the profile. Sizing is free — it never spends credits; only revealing contact details is metered.

    Annotations the server declares

    • title: Refresh ICP audience size
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: true
  • resume_autopilot_runResumes a paused autopilot run.No findings

    Why it is in this group

    • The name "resume_autopilot_run" says "run": a word for a change.
    Description, as the server wrote it
    Resumes a paused autopilot run. A run paused before approval goes back to awaiting approval — resuming can never skip the gate. A failed run resumes from the stage it failed at once the cause (e.g. insufficient credits) is fixed.

    Annotations the server declares

    • title: Autopilot: resume a run
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: false
  • set_billing_profileCreates or replaces the workspace's registrant and postal contact details.Can send data outNo findings

    Why it is in this group

    • The name "set_billing_profile" says "set": a word for a change.

    What it can reach

    • outboundThe input takes a destination (input.phone): what the tool writes goes there.
    Description, as the server wrote it
    Creates or replaces the workspace's registrant and postal contact details. Domain registration files these details with the registrar, so every field except addressLineTwo is required; a done-for-you order uses them when it registers domains. Idempotent: sending the same body twice leaves the same state.

    Annotations the server declares

    • title: Set workspace billing profile
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • set_primary_icpPromotes an Ideal Customer Profile to the workspace's active one, demoting any existing primary.No findings

    Why it is in this group

    • The name "set_primary_icp" says "set": a word for a change.
    Description, as the server wrote it
    Promotes an Ideal Customer Profile to the workspace's active one, demoting any existing primary. At most one profile is primary per workspace.

    Annotations the server declares

    • title: Set primary ICP
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • start_autopilot_runStarts an autopilot run for a company website: the AI builds an ICP, writes a sequence and checks that prospects match, then STOPS at an approval gate.Reads text others writeNo findings

    Why it is in this group

    • The name "start_autopilot_run" says "start": a word for a change.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Starts an autopilot run for a company website: the AI builds an ICP, writes a sequence and checks that prospects match, then STOPS at an approval gate. Nothing is revealed, charged or sent and no sending accounts are bought until the run is approved; approval then reveals the first batch at 1 credit ($0.033 at list) per prospect, up to targetProspects and at most 2,000. The call is refused when the wallet cannot cover that batch. When budgetUsd is given, the sized budget plan is snapshotted on the run and returned in this response, the only response that carries it, with credits valued at list, and targetProspects defaults to the plan's monthly prospects. Without budgetUsd the run can never buy sending accounts, so the workspace needs one connected before approval or the run fails. Requires an active subscription (any plan).

    Annotations the server declares

    • title: Autopilot: start a run
    • readOnlyHint: false
    • destructiveHint: false
    • openWorldHint: true
  • update_blocklist_entryChanges one suppression entry's blocked value, its scope, or both; provide at least one.No findings

    Why it is in this group

    • The name "update_blocklist_entry" says "update": a word for a change.
    Description, as the server wrote it
    Changes one suppression entry's blocked value, its scope, or both; provide at least one. A value containing @ is stored as an email address, otherwise as a domain, so an entry can be converted between the two. Moving an entry to or from the account-wide list ("global") requires the main workspace's key and moves the entry onto the main workspace.

    Annotations the server declares

    • title: Update blocklist entry
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • update_campaignUpdates campaign properties such as name, emoji, timezone, sending limits, the minimum email account health score and deliverability settings.No findings

    Why it is in this group

    • The name "update_campaign" says "update": a word for a change.
    Description, as the server wrote it
    Updates campaign properties such as name, emoji, timezone, sending limits, the minimum email account health score and deliverability settings. Only the provided fields are changed. A new minimum health score re-plans a running campaign at once.

    Annotations the server declares

    • title: Update campaign
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • update_campaign_scheduleUpdates a campaign's sending schedule: timezone, sending days, daily time window and frequency.No findings

    Why it is in this group

    • The name "update_campaign_schedule" says "update": a word for a change.
    Description, as the server wrote it
    Updates a campaign's sending schedule: timezone, sending days, daily time window and frequency. everySchedule can be left out: the campaign keeps its current frequency, or 5 minutes (the app's default) if it has none.

    Annotations the server declares

    • title: Update campaign schedule
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • update_icpApplies a partial edit to an Ideal Customer Profile: omitted fields are left unchanged, and sending an empty list clears that list.No findings

    Why it is in this group

    • The name "update_icp" says "update": a word for a change.
    Description, as the server wrote it
    Applies a partial edit to an Ideal Customer Profile: omitted fields are left unchanged, and sending an empty list clears that list. Any edit marks the profile as human-authored.

    Annotations the server declares

    • title: Update ICP
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • update_leadUpdates a lead's contact details.Can send data outNo findings

    Why it is in this group

    • The name "update_lead" says "update": a word for a change.

    What it can reach

    • outboundThe input takes a destination (input.phone): what the tool writes goes there.
    Description, as the server wrote it
    Updates a lead's contact details. Only the provided fields are changed.

    Annotations the server declares

    • title: Update lead
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • update_lead_categorySets or corrects a lead's category (tag), typically to fix an AI misclassification of a reply.No findings

    Why it is in this group

    • The name "update_lead_category" says "update": a word for a change.
    Description, as the server wrote it
    Sets or corrects a lead's category (tag), typically to fix an AI misclassification of a reply. Setting meeting_booked behaves exactly like marking a meeting as booked. Changing the category away from meeting_booked does NOT clear the booked-meeting mark: that mark is removed only by unmarking the meeting. A real change fires the LeadCategoryUpdate webhook; setting the value the lead already has is a no-op.

    Annotations the server declares

    • title: Update lead category
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • update_sender_emailUpdates a connected sending email account's display name, signature or daily sending limits.Can send data outNo findings

    Why it is in this group

    • The name "update_sender_email" says "update": a word for a change.

    What it can reach

    • outboundThe name "update_sender_email" says "update" and "email": it writes something other people read.
    Description, as the server wrote it
    Updates a connected sending email account's display name, signature or daily sending limits. Only the provided fields are changed. Warm-up settings are separate and are not changed here. The returned account includes its healthScore (0-100, higher is better; null while warm-up is off or before 20 warm-up emails were checked).

    Annotations the server declares

    • title: Update sender email
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • update_sender_email_warmupSwitches warm-up on or off and changes the warm-up ramp for up to 100 mailboxes in one call, applying the same settings to each.Can send data outNo findings

    Why it is in this group

    • The name "update_sender_email_warmup" says "update": a word for a change.

    What it can reach

    • outboundThe name "update_sender_email_warmup" says "update" and "email": it writes something other people read.
    Description, as the server wrote it
    Switches warm-up on or off and changes the warm-up ramp for up to 100 mailboxes in one call, applying the same settings to each. Only the settings provided change. Warm-up sends startLimit emails on the first day and adds increaseBy each sending day until it reaches capLimit a day; this is separate from campaign sending limits. A mailbox that has never warmed needs enabled: true to take any other setting, and is then enrolled with the defaults (2 a day, 2 more each day, up to 10 a day, weekdays only, UTC) plus the settings given. Raising capLimit on a mailbox that is already warming lifts today's target without restarting its ramp. Each mailbox succeeds or fails on its own: the result lists the updated mailboxes' settings, each with its healthScore (0-100, higher is better; null while warm-up is off or before 20 warm-up emails were checked), under updated, and any failures with the reason under failed.

    Annotations the server declares

    • title: Update sender email warm-up
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • update_webhookUpdates a registered webhook's name, URL, event type or enabled state.Can send data outNo findings

    Why it is in this group

    • The name "update_webhook" says "update": a word for a change.

    What it can reach

    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    Updates a registered webhook's name, URL, event type or enabled state. Only the provided fields are changed. Disabling a webhook stops deliveries without deleting it. API reference: https://www.emailchaser.com/docs/api-reference/webhooks

    Annotations the server declares

    • title: Update webhook
    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
Only reads 47Show 47 read-only tools
  • get_audience_sizeCounts how many prospects match ad-hoc targeting criteria (titles, seniorities, industries, company sizes, locations, keywords) without saving anything.Reads text others writeNo findings

    Why it is in this group

    • The name "get_audience_size" says "get": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Counts how many prospects match ad-hoc targeting criteria (titles, seniorities, industries, company sizes, locations, keywords) without saving anything. Free — searching never spends credits; only revealing contact details is metered. Use it to size an audience and validate targeting. A read-only key may call it.

    Annotations the server declares

    • title: Get audience size
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
  • get_autopilot_runReturns one autopilot run including its full audit trail.No findings

    Why it is in this group

    • The name "get_autopilot_run" says "get": a word for reading.
    Description, as the server wrote it
    Returns one autopilot run including its full audit trail. It does not return the budget plan: that is returned only in the response that started the run, when budgetUsd was given. Use it to check progress.

    Annotations the server declares

    • title: Autopilot: get a run
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_billing_profileReturns the registrant and postal contact details held for the workspace: what is filed with the registrar when a done-for-you order buys a domain, and the postal address a CAN-SPA…Reads private dataNo findings

    Why it is in this group

    • The name "get_billing_profile" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "return" and "contact": it reads contacts.
    Description, as the server wrote it
    Returns the registrant and postal contact details held for the workspace: what is filed with the registrar when a done-for-you order buys a domain, and the postal address a CAN-SPAM footer must carry. Errors with 404 until the profile has been set.

    Annotations the server declares

    • title: Get workspace billing profile
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_blocklist_entryReturns one suppression entry by ID: the blocked domain or email address, when it was added and which list it belongs to.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_blocklist_entry" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "return" and "email": it reads mail.
    • untrusted-inputThe description's first sentence says "return" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Returns one suppression entry by ID: the blocked domain or email address, when it was added and which list it belongs to. Entries inherited from the account-wide list are visible from a sub-workspace too and report scope "global".

    Annotations the server declares

    • title: Get blocklist entry
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_campaignRetrieves a campaign by ID, including settings, sending schedule, stats and per-status email counts.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_campaign" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "retrieve" and "email": it reads mail.
    • untrusted-inputThe description's first sentence says "retrieve" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Retrieves a campaign by ID, including settings, sending schedule, stats and per-status email counts. minimumHealth lists each connected email account on the campaign with its healthScore and whether the campaign's minimum health score (settings.minimumHealthScore) holds it back, with the reason; when every account is held back, allHeldBack is true and message says the campaign is sending nothing.

    Annotations the server declares

    • title: Get campaign
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_campaign_sequenceReturns the ordered emails that make up a campaign's sequence, including any A/B variants.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_campaign_sequence" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "return" and "email": it reads mail.
    • untrusted-inputThe description's first sentence says "return" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Returns the ordered emails that make up a campaign's sequence, including any A/B variants. delayDays on each step is counted from the previous step, and the first step is always 0.

    Annotations the server declares

    • title: Get campaign sequence
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_campaign_statsReturns a campaign's performance in one call: totals (sent, replied, positive, bounced, meetings), per-A/B-variant results, per-sequence-step results and a daily time series (UTC d…Reads private dataNo findings

    Why it is in this group

    • The name "get_campaign_stats" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "return" and "meeting": it reads calendars.
    Description, as the server wrote it
    Returns a campaign's performance in one call: totals (sent, replied, positive, bounced, meetings), per-A/B-variant results, per-sequence-step results and a daily time series (UTC days). Use it to judge how a campaign is doing or to compare variants. There is no opens metric anywhere in the response: Emailchaser does not track opens, so its absence is deliberate, not an omission.

    Annotations the server declares

    • title: Get campaign stats
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_credit_balanceReturns the workspace's credit wallet: available, reserved, lifetime granted and used, and the monthly grant.No findings

    Why it is in this group

    • The name "get_credit_balance" says "get": a word for reading.
    Description, as the server wrote it
    Returns the workspace's credit wallet: available, reserved, lifetime granted and used, and the monthly grant. Credits pay for prospect reveals and AI work (list price $0.033 each). A new workspace sees zeros, not an error. unlimited is true when Emailchaser has made this workspace's credits free: every credit action then goes through without using available and is never refused for lack of credits, so there is no need to check the balance or buy credits.

    Annotations the server declares

    • title: Get credit balance
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_deliverability_insightsWhat is wrong with deliverability across the workspace right now, worst first, each with the action that fixes it.No findings

    Why it is in this group

    • The name "get_deliverability_insights" says "get": a word for reading.
    Description, as the server wrote it
    What is wrong with deliverability across the workspace right now, worst first, each with the action that fixes it. Combines the inbox placement measured over the window with the latest health check of every connected mailbox: authentication records, blacklist listings and measured placement per mailbox. The best first call when asked 'why are we landing in spam'.

    Annotations the server declares

    • title: Get deliverability insights
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_dfy_orderReturns one done-for-you order: status (created, pending_approval, processing, completed, failed, partially_completed, canceled), cost breakdown and any failure reason.No findings

    Why it is in this group

    • The name "get_dfy_order" says "get": a word for reading.
    Description, as the server wrote it
    Returns one done-for-you order: status (created, pending_approval, processing, completed, failed, partially_completed, canceled), cost breakdown and any failure reason. Use it to track an order's provisioning.

    Annotations the server declares

    • title: Get done-for-you order
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_email_verification_jobReturns one verification job: its state, per-verdict counts and what it has billed.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_email_verification_job" says "get": a word for reading.

    What it can reach

    • private-dataThe name "get_email_verification_job" says "get" and "email": it reads mail.
    • untrusted-inputThe name "get_email_verification_job" says "get" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Returns one verification job: its state, per-verdict counts and what it has billed. Poll until state is done. An errorCode of monthly_limit_reached means the allowance ran out and the remaining addresses came back unknown rather than missing — they are still in the results, marked, and were not billed.

    Annotations the server declares

    • title: Get an email verification job
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_email_verification_ratesReturns what one email verification costs right now, read from the billing plan the meter actually charges against.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_email_verification_rates" says "get": a word for reading.

    What it can reach

    • private-dataThe name "get_email_verification_rates" says "get" and "email": it reads mail.
    • private-dataThe description's first sentence says "return" and "email": it reads mail.
    • untrusted-inputThe name "get_email_verification_rates" says "get" and "email": it reads inbound email, which other people can write.
    • untrusted-inputThe description's first sentence says "return" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Returns what one email verification costs right now, read from the billing plan the meter actually charges against. The price is read live, never a fixed figure. Every check runs a waterfall of up to two providers and each one that returns a verdict is one billable verification credit, so an address costs ONE credit when the first provider rejects it outright and TWO when both answer. Budget against perAddressCeilingUsd, which is the common case, not perAddressFloorUsd.

    Annotations the server declares

    • title: Get the email verification rate
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_icpRetrieves one Ideal Customer Profile by ID, including its targeting criteria, whether it is AI-generated or human-edited, primary status and estimated audience size.No findings

    Why it is in this group

    • The name "get_icp" says "get": a word for reading.
    Description, as the server wrote it
    Retrieves one Ideal Customer Profile by ID, including its targeting criteria, whether it is AI-generated or human-edited, primary status and estimated audience size.

    Annotations the server declares

    • title: Get ICP
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_inbox_placement_runReturns one inbox placement run with its complete report: the headline inbox / spam / promotions / missing split, the breakdown per mailbox provider and per sending mailbox, and th…Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_inbox_placement_run" says "get": a word for reading.

    What it can reach

    • private-dataThe name "get_inbox_placement_run" says "get" and "inbox": it reads mail.
    • private-dataThe description's first sentence says "return" and "inbox": it reads mail.
    • untrusted-inputThe name "get_inbox_placement_run" says "get" and "inbox": it reads inbound email, which other people can write.
    • untrusted-inputThe description's first sentence says "return" and "inbox": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Returns one inbox placement run with its complete report: the headline inbox / spam / promotions / missing split, the breakdown per mailbox provider and per sending mailbox, and the content spam rules the copy triggered. 'Promotions' rolls up every Gmail category tab. 'Missing' means the probe was never found in any folder, which usually indicates a silent block: a worse problem than spam, with a different fix. spamScore is in tenths of a point (47 means 4.7) and -1 means not scored.

    Annotations the server declares

    • title: Get inbox placement run
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_inbox_placement_stats_by_dateRolls completed inbox placement runs up by UTC calendar day over a window, for a trend.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_inbox_placement_stats_by_date" says "get": a word for reading.

    What it can reach

    • private-dataThe name "get_inbox_placement_stats_by_date" says "get" and "inbox": it reads mail.
    • private-dataThe description's first sentence says "inbox": it reads mail.
    • untrusted-inputThe name "get_inbox_placement_stats_by_date" says "get" and "inbox": it reads inbound email, which other people can write.
    • untrusted-inputThe description's first sentence says "inbox": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Rolls completed inbox placement runs up by UTC calendar day over a window, for a trend. Defaults to the last 30 days. Days on which nothing ran are ABSENT from the series rather than zero: never report a missing day as 0% placement, it means nobody ran a test that day.

    Annotations the server declares

    • title: Get inbox placement stats by date
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_leadRetrieves a lead by ID, including contact details and company information.Reads private dataNo findings

    Why it is in this group

    • The name "get_lead" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "retrieve" and "contact": it reads contacts.
    Description, as the server wrote it
    Retrieves a lead by ID, including contact details and company information.

    Annotations the server declares

    • title: Get lead
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_lead_conversationReturns a lead's full email thread in chronological order: outbound emails (sent, scheduled and unsent drafts) plus inbound replies, each with direction and status.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_lead_conversation" says "get": a word for reading.

    What it can reach

    • private-dataThe name "get_lead_conversation" says "get" and "conversation": it reads messages.
    • private-dataThe description's first sentence says "return" and "email": it reads mail.
    • untrusted-inputThe name "get_lead_conversation" says "get" and "conversation": it reads messages, which other people can write.
    • untrusted-inputThe description's first sentence says "return" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Returns a lead's full email thread in chronological order: outbound emails (sent, scheduled and unsent drafts) plus inbound replies, each with direction and status. AI reply drafts awaiting review are marked isDraft. Every inbound reply in the thread, with or without an AI draft, can be answered with your own message that uses that reply's id. Use it to read the back-and-forth with one lead before deciding what to do next.

    Annotations the server declares

    • title: Get lead conversation
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_lead_finder_filtersReturns the values Lead Finder's list filters accept (seniorities, jobFunctions, companySizes, revenue, industries, countries, headquartersCountries, regions, continents), the limi…No findings

    Why it is in this group

    • The name "get_lead_finder_filters" says "get": a word for reading.
    Description, as the server wrote it
    Returns the values Lead Finder's list filters accept (seniorities, jobFunctions, companySizes, revenue, industries, countries, headquartersCountries, regions, continents), the limits that apply to this workspace (page sizes, deepest page, people per add, first-N count, imports running at once, and the daily browsing allowance with what is left today), and creditsPerProspect, what adding one person to a campaign costs. Free, and makes no call to the data provider. Filter values in a search must match these values exactly.

    Annotations the server declares

    • title: Lead Finder: filter values and limits
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_lead_finder_importReturns one Lead Finder add's progress: status (running, completed, failed or canceled), how many people were added and skipped and why (skipReasons), what email verification said,…Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_lead_finder_import" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "return" and "email": it reads mail.
    • untrusted-inputThe description's first sentence says "return" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Returns one Lead Finder add's progress: status (running, completed, failed or canceled), how many people were added and skipped and why (skipReasons), what email verification said, and credits spent. added, creditsSpent and status can still change until finishedAt is set, so poll it until it is. A completed add's endReason is all_selected, requested_reached, audience_exhausted or fetch_limit, and a canceled one's is canceled. A failed add can still have added and charged people (see added and creditsSpent); lastError says why it stopped: insufficient_credits, contact_cap_reached, results_expired, budget_exhausted, provider_blocked, invalid_filters, campaign_not_found or internal. A new filters add to finish a partial one starts from the top again and counts the people already added toward its fetch limit (five people per one requested, at least 1,000), so after a large add, pick the missing people by ref from the search results instead.

    Annotations the server declares

    • title: Lead Finder: add progress
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_primary_icpReturns the workspace's active (primary) Ideal Customer Profile — the profile the workspace currently targets; at most one is primary.No findings

    Why it is in this group

    • The name "get_primary_icp" says "get": a word for reading.
    Description, as the server wrote it
    Returns the workspace's active (primary) Ideal Customer Profile — the profile the workspace currently targets; at most one is primary. Errors with 404 when none is set.

    Annotations the server declares

    • title: Get primary ICP
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_sender_emailRetrieves a connected sending email account by ID, including its health score.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_sender_email" says "get": a word for reading.

    What it can reach

    • private-dataThe name "get_sender_email" says "get" and "email": it reads mail.
    • private-dataThe description's first sentence says "retrieve" and "email": it reads mail.
    • untrusted-inputThe name "get_sender_email" says "get" and "email": it reads inbound email, which other people can write.
    • untrusted-inputThe description's first sentence says "retrieve" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Retrieves a connected sending email account by ID, including its health score. healthScore is the account's health score: 0-100, higher is better, the share of its warm-up emails over the last 7 full days that landed in the inbox rather than spam. It moves daily as warm-up emails land in the inbox (up) or in spam (down), and is null while warm-up is off or before 20 warm-up emails were checked.

    Annotations the server declares

    • title: Get sender email
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_sender_email_dnsChecks the sender domain's SPF, DKIM, DMARC and MX records and reports each as OK, WARNING, MISSING or ERROR with human-readable issues (the same checks behind the app's DNS vitals…Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_sender_email_dns" says "get": a word for reading.

    What it can reach

    • private-dataThe name "get_sender_email_dns" says "get" and "email": it reads mail.
    • private-dataThe description's first sentence says "mailbox": it reads mail.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • untrusted-inputThe name "get_sender_email_dns" says "get" and "email": it reads inbound email, which other people can write.
    • untrusted-inputThe description's first sentence says "issue": it reads issues, which other people can write.
    Description, as the server wrote it
    Checks the sender domain's SPF, DKIM, DMARC and MX records and reports each as OK, WARNING, MISSING or ERROR with human-readable issues (the same checks behind the app's DNS vitals view), plus the mailbox's warm-up status: whether it is enrolled, today's ramp target, the daily cap and its healthScore (0-100, higher is better, the share of its warm-up emails over the last 7 full days that landed in the inbox rather than spam). Use it to diagnose deliverability. DNS lookups run live at request time, so the call can take a few seconds but never returns stale records.

    Annotations the server declares

    • title: Get sender email DNS vitals
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
  • get_sender_email_warmupReads a mailbox's warm-up settings: whether warm-up is on, the ramp (startLimit warm-up emails on the first day, increaseBy more each sending day, up to capLimit a day), weekdays-o…Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_sender_email_warmup" says "get": a word for reading.

    What it can reach

    • private-dataThe name "get_sender_email_warmup" says "get" and "email": it reads mail.
    • private-dataThe description's first sentence says "read" and "mailbox": it reads mail.
    • untrusted-inputThe name "get_sender_email_warmup" says "get" and "email": it reads inbound email, which other people can write.
    • untrusted-inputThe description's first sentence says "read" and "mailbox": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Reads a mailbox's warm-up settings: whether warm-up is on, the ramp (startLimit warm-up emails on the first day, increaseBy more each sending day, up to capLimit a day), weekdays-only, timezone, today's ramp target (currentPerDay) and the account's healthScore (0-100, higher is better, the share of its warm-up emails over the last 7 full days that landed in the inbox rather than spam; null while warm-up is off or before 20 were checked). Warm-up volume is separate from the mailbox's campaign sending limits. When configured is false the mailbox has never warmed, and the values shown are the defaults that switching warm-up on would use.

    Annotations the server declares

    • title: Get sender email warm-up settings
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_sender_reputationReturns the latest health check for every connected mailbox: SPF, DKIM, DMARC and MX status (OK, WARNING, MISSING or ERROR), blacklist listings with their delisting links, measured…Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_sender_reputation" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "return" and "mailbox": it reads mail.
    • untrusted-inputThe description's first sentence says "return" and "mailbox": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Returns the latest health check for every connected mailbox: SPF, DKIM, DMARC and MX status (OK, WARNING, MISSING or ERROR), blacklist listings with their delisting links, measured inbox placement over recent runs and a combined 0-100 health score. blacklistsChecked is reported next to blacklistsListed so the count is never read as a total. A placementScore or healthScore of -1 means not measured, not zero.

    Annotations the server declares

    • title: Get sender reputation and mailbox health
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • get_workspace_detailsReturns the workspace behind the API key: campaign counts broken down by status and the list of members.No findings

    Why it is in this group

    • The name "get_workspace_details" says "get": a word for reading.
    Description, as the server wrote it
    Returns the workspace behind the API key: campaign counts broken down by status and the list of members. Useful as a first call to confirm which workspace a key belongs to.

    Annotations the server declares

    • title: Get workspace details
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_autopilot_runsLists the workspace's autopilot runs, newest first, with status (pending, building_icp, writing_sequence, sourcing_prospects, awaiting_approval, running, paused, completed, failed)…No findings

    Why it is in this group

    • The name "list_autopilot_runs" says "list": a word for reading.
    Description, as the server wrote it
    Lists the workspace's autopilot runs, newest first, with status (pending, building_icp, writing_sequence, sourcing_prospects, awaiting_approval, running, paused, completed, failed) and linked ICP/campaign IDs.

    Annotations the server declares

    • title: Autopilot: list runs
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_blocklist_entriesReturns the suppression entries (blocked domains and email addresses) that apply to this workspace, newest first.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_blocklist_entries" says "list": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "return" and "email": it reads mail.
    • untrusted-inputThe description's first sentence says "return" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Returns the suppression entries (blocked domains and email addresses) that apply to this workspace, newest first. Blocked entries are never emailed by any campaign. By default that spans both lists: this workspace's own entries and the account-wide ones inherited from the main workspace. Each entry reports which list it came from.

    Annotations the server declares

    • title: List blocklist entries
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_campaignsLists cold email campaigns in the workspace with their status and stats (sent, replies, bounces).Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_campaigns" says "list": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "list" and "email": it reads mail.
    • private-dataThe input names a folder (input.folderId), and the tool reads: it reads files.
    • untrusted-inputThe description's first sentence says "list" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Lists cold email campaigns in the workspace with their status and stats (sent, replies, bounces). Supports pagination and filtering by status or folder.

    Annotations the server declares

    • title: List campaigns
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_credit_transactionsLists the workspace's credit ledger, newest first: every grant, spend, reservation and refund with the balance after each entry.No findings

    Why it is in this group

    • The name "list_credit_transactions" says "list": a word for reading.
    Description, as the server wrote it
    Lists the workspace's credit ledger, newest first: every grant, spend, reservation and refund with the balance after each entry. Filter by movement kind, reason or a time window. Use it to see where credits went, or to check for a stripe_topup entry after a credit purchase whose outcome was unconfirmed. An entry with free set to true was written while the workspace's credits were free: its amount is what the action would have cost, and no credits moved.

    Annotations the server declares

    • title: List credit transactions
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_dfy_ordersLists the workspace's done-for-you orders, newest first, each with status, cost breakdown and the ordered domains and mailboxes.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_dfy_orders" says "list": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "list" and "mailbox": it reads mail.
    • untrusted-inputThe description's first sentence says "list" and "mailbox": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Lists the workspace's done-for-you orders, newest first, each with status, cost breakdown and the ordered domains and mailboxes. Internal billing and provider records are never exposed.

    Annotations the server declares

    • title: List done-for-you orders
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_email_verification_jobsLists the workspace's standalone email verification jobs, newest first, with live counts and what each has billed so far (meterEvents and spentUsd).Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_email_verification_jobs" says "list": a word for reading.

    What it can reach

    • private-dataThe name "list_email_verification_jobs" says "list" and "email": it reads mail.
    • private-dataThe description's first sentence says "list" and "email": it reads mail.
    • untrusted-inputThe name "list_email_verification_jobs" says "list" and "email": it reads inbound email, which other people can write.
    • untrusted-inputThe description's first sentence says "list" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Lists the workspace's standalone email verification jobs, newest first, with live counts and what each has billed so far (meterEvents and spentUsd).

    Annotations the server declares

    • title: List email verification jobs
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_email_verification_resultsReturns one row per address with its verdict: valid, catchall_validated (the domain accepts everything, so delivery is likely but not proven), invalid, or unknown (no verdict, and …Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_email_verification_results" says "list": a word for reading.

    What it can reach

    • private-dataThe name "list_email_verification_results" says "list" and "email": it reads mail.
    • untrusted-inputThe name "list_email_verification_results" says "list" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Returns one row per address with its verdict: valid, catchall_validated (the domain accepts everything, so delivery is likely but not proven), invalid, or unknown (no verdict, and not billed). Each row also carries what each provider said and how many verification credits it cost. Filter with result, where 'deliverable' means valid plus catch-all, which is what a campaign will actually send to.

    Annotations the server declares

    • title: List an email verification job's results
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_icpsLists the workspace's stored Ideal Customer Profiles (ICPs), newest first, with their targeting criteria, whether each is AI-generated or human-edited, and the estimated audience s…No findings

    Why it is in this group

    • The name "list_icps" says "list": a word for reading.
    Description, as the server wrote it
    Lists the workspace's stored Ideal Customer Profiles (ICPs), newest first, with their targeting criteria, whether each is AI-generated or human-edited, and the estimated audience size.

    Annotations the server declares

    • title: List ICPs
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_inbox_placement_run_resultsReturns one row per probe email of a run: which of your mailboxes sent it, which seed mailbox received it, where it landed and how many seconds it took to arrive.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_inbox_placement_run_results" says "list": a word for reading.

    What it can reach

    • private-dataThe name "list_inbox_placement_run_results" says "list" and "inbox": it reads mail.
    • private-dataThe description's first sentence says "return" and "email": it reads mail.
    • untrusted-inputThe name "list_inbox_placement_run_results" says "list" and "inbox": it reads inbound email, which other people can write.
    • untrusted-inputThe description's first sentence says "return" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Returns one row per probe email of a run: which of your mailboxes sent it, which seed mailbox received it, where it landed and how many seconds it took to arrive. deliverySeconds is worth watching on its own: greylisting and throttling show up there before they show up in a placement number.

    Annotations the server declares

    • title: List inbox placement run probes
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_inbox_placement_runsLists inbox placement runs, newest first, optionally only those of one test.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_inbox_placement_runs" says "list": a word for reading.

    What it can reach

    • private-dataThe name "list_inbox_placement_runs" says "list" and "inbox": it reads mail.
    • private-dataThe description's first sentence says "list" and "inbox": it reads mail.
    • untrusted-inputThe name "list_inbox_placement_runs" says "list" and "inbox": it reads inbound email, which other people can write.
    • untrusted-inputThe description's first sentence says "list" and "inbox": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Lists inbox placement runs, newest first, optionally only those of one test. Each run counts how many probe emails landed in the inbox, in spam, in a Gmail category tab (promotions) or nowhere (missing, which usually means a silent block). Counters are only final once status is completed. Any rate of -1 means NOT MEASURED yet and must never be reported as 0%.

    Annotations the server declares

    • title: List inbox placement runs
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_inbox_placement_testsLists every inbox placement test defined in the workspace, newest first.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_inbox_placement_tests" says "list": a word for reading.

    What it can reach

    • private-dataThe name "list_inbox_placement_tests" says "list" and "inbox": it reads mail.
    • private-dataThe description's first sentence says "list" and "inbox": it reads mail.
    • untrusted-inputThe name "list_inbox_placement_tests" says "list" and "inbox": it reads inbound email, which other people can write.
    • untrusted-inputThe description's first sentence says "list" and "inbox": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Lists every inbox placement test defined in the workspace, newest first. A test is the definition: the email content, which mailboxes send it and, for a recurring test, how often it runs. Each execution of a test is a run. Inbox placement is a paid add-on: a workspace without it gets an error, not an empty list.

    Annotations the server declares

    • title: List inbox placement tests
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_instantly_accountsFetches the sending accounts of an Instantly.ai workspace using your Instantly API key, together with a pre-filled CSV for Emailchaser's bulk IMAP/SMTP connection flow.Reads text others writeNo findings

    Why it is in this group

    • The name "list_instantly_accounts" says "list": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Fetches the sending accounts of an Instantly.ai workspace using your Instantly API key, together with a pre-filled CSV for Emailchaser's bulk IMAP/SMTP connection flow. A read-only preview before an import: nothing is written to the workspace. Mailbox passwords cannot be exported from any provider, so each account is either reconnected via Google/Microsoft OAuth in the app or bulk-uploaded with app passwords using the CSV. The backend treats all non-GET calls as writes, so this needs a read & write API key.

    Annotations the server declares

    • title: List Instantly sending accounts
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
  • list_lead_finder_importsLists the workspace's Lead Finder adds (people added to campaigns from the contact database), newest first, each with its progress: requested, added, skipped and why, what email ve…Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_lead_finder_imports" says "list": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "list" and "contact": it reads contacts.
    • untrusted-inputThe description's first sentence says "list" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Lists the workspace's Lead Finder adds (people added to campaigns from the contact database), newest first, each with its progress: requested, added, skipped and why, what email verification said, and credits spent.

    Annotations the server declares

    • title: Lead Finder: list adds
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_leadsLists the leads in the workspace, newest first, 20 per page.No findings

    Why it is in this group

    • The name "list_leads" says "list": a word for reading.
    Description, as the server wrote it
    Lists the leads in the workspace, newest first, 20 per page. Filter by exact email to look up a single lead, or by campaign to list only that campaign's leads. This is how you find the ID of a lead you added earlier.

    Annotations the server declares

    • title: List leads
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_repliesLists inbound emails (replies received from prospects) in the workspace, newest first, 20 per page.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_replies" says "list": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "list" and "email": it reads mail.
    • untrusted-inputThe name "list_replies" says "list" and "reply": it reads comments, which other people can write.
    • untrusted-inputThe description's first sentence says "list" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Lists inbound emails (replies received from prospects) in the workspace, newest first, 20 per page. Filter by AI response category, campaign, lead or a time floor. responseCategory is null while AI categorization is still pending. This tool only lists. Any reply can be answered with your own message, including ones with no AI draft, such as wrong_person or uncategorized; a reply that has an AI draft can also be answered by editing and sending that draft.

    Annotations the server declares

    • title: List replies
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_reply_draftsLists AI-suggested reply drafts awaiting review, newest first, 20 per page.Reads text others writeNo findings

    Why it is in this group

    • The name "list_reply_drafts" says "list": a word for reading.

    What it can reach

    • untrusted-inputThe name "list_reply_drafts" says "list" and "reply": it reads comments, which other people can write.
    • untrusted-inputThe description's first sentence says "list" and "reply": it reads comments, which other people can write.
    Description, as the server wrote it
    Lists AI-suggested reply drafts awaiting review, newest first, 20 per page. Each draft answers the inbound reply referenced by inReplyToEmailId. Nothing here has been sent: a draft stays a draft until it is sent. A draft's subject and body can be edited before it goes out.

    Annotations the server declares

    • title: List AI reply drafts
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_sender_emailsLists the sending email accounts in the workspace, 20 per page, with their connection status, daily limits and health score.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "list_sender_emails" says "list": a word for reading.

    What it can reach

    • private-dataThe name "list_sender_emails" says "list" and "email": it reads mail.
    • private-dataThe description's first sentence says "list" and "email": it reads mail.
    • untrusted-inputThe name "list_sender_emails" says "list" and "email": it reads inbound email, which other people can write.
    • untrusted-inputThe description's first sentence says "list" and "email": it reads inbound email, which other people can write.
    Description, as the server wrote it
    Lists the sending email accounts in the workspace, 20 per page, with their connection status, daily limits and health score. Filter to one campaign's senders, to connected or disconnected mailboxes only, or to one workspace on the account. healthScore is the account's health score: 0-100, higher is better, the share of its warm-up emails over the last 7 full days that landed in the inbox rather than spam. It moves daily as warm-up emails land in the inbox (up) or in spam (down), and is null while warm-up is off or before 20 warm-up emails were checked. Use it to pick the accounts to add to a campaign.

    Annotations the server declares

    • title: List sender emails
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_webhooksLists the webhook endpoints registered for the workspace, with their event type and status.No findings

    Why it is in this group

    • The name "list_webhooks" says "list": a word for reading.
    Description, as the server wrote it
    Lists the webhook endpoints registered for the workspace, with their event type and status.

    Annotations the server declares

    • title: List webhooks
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • list_workspacesLists the workspaces owned by the account behind the API key: the main workspace and its sub-workspaces, each with id, name, icon and creation date.1 finding

    Why it is in this group

    • The name "list_workspaces" says "list": a word for reading.

    Findings

    • Warn

      The description contains "Every other tool", wording about other tools or servers.

      Matched text

      description: "...rks the workspace this key is bound to. Every other tool acts on the current workspace only; to..."

      What this check looks for
    Description, as the server wrote it
    Lists the workspaces owned by the account behind the API key: the main workspace and its sub-workspaces, each with id, name, icon and creation date. isCurrent marks the workspace this key is bound to. Every other tool acts on the current workspace only; to work in another workspace, connect with that workspace's own key.

    Annotations the server declares

    • title: List workspaces
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: false
  • search_dfy_domainsChecks availability and pricing of sending domains derived from a brand name (e.g.Reads text others writeNo findings

    Why it is in this group

    • The name "search_dfy_domains" says "search": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Checks availability and pricing of sending domains derived from a brand name (e.g. 'acme' yields acme-mail.com and similar). Only .com and .org are supported. Free to call. Returns the exact price of each available domain.

    Annotations the server declares

    • title: Search available sending domains
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true
  • search_lead_finderSearches Emailchaser's B2B contact database (Lead Finder) with explicit filters and returns one page of matching people, masked: first name, last initial, title, seniority, job fun…Reads private dataReads text others write1 finding

    Why it is in this group

    • The name "search_lead_finder" says "search": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "search" and "contact": it reads contacts.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Info

      The description is 2,273 characters long, over the 2,000 this check looks for.

      Matched text

      description ends: "...alse, and adding them is skipped for free too. A read-only key may use this tool."

      What this check looks for
    Description, as the server wrote it
    Searches Emailchaser's B2B contact database (Lead Finder) with explicit filters and returns one page of matching people, masked: first name, last initial, title, seniority, job function, company name, industry, size and revenue band, and location. Email addresses, domains, LinkedIn URLs and phone numbers are never shown; contact details are revealed only when people are added to a campaign. Spends no credits, but every result row uses the account's daily browsing allowance (rowsLeftToday; 2,000 rows a day by default, 250 on a trial). An account may start 20 searches a minute. Searches that find nobody also draw on a bucket of 120 that refills one every 30 seconds; while it is empty every new search is refused for a moment (rate_limited, reason empty_searches, with retryAfterSeconds). The same page asked for again within 10 minutes costs nothing. total is the exact audience size only when totalIsExact is true; otherwise it is a lower bound (often 50,000). If status is running or totalStatus is pending, total, hasMore and maxPage are recomputed when the search is read again by its searchId. The exact count is fetched in the background and can take up to about a minute, so totalStatus can read pending (and totalIsExact false) on several polls in a row: that is normal, not an error. The count is settled once totalIsExact is true or totalStatus is done (exact) or failed (the count could not be fetched, so total stays a lower bound); each read of the search waits about six seconds, so around ten reads cover it. The people already returned are final and usable the whole time: show them, and add them with their refs, without waiting for the count. status failed means the search did not run, and error says why (rate_limited, timeout, provider_blocked, budget_exhausted, invalid_filters, expired or internal): it is not an empty audience, so try again later. Each result's ref identifies that person when adding them to a campaign, valid for 60 minutes after its page was last shown. inWorkspace means Lead Finder added that person to this workspace and their lead is still there, so adding them again is skipped for free; someone who is a lead from another source shows false, and adding them is skipped for free too. A read-only key may use this tool.

    Annotations the server declares

    • title: Lead Finder: search the contact database (free)
    • readOnlyHint: true
    • destructiveHint: false
    • openWorldHint: true

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

16008 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 12 outbound tools breaks every one.

deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__emailchaser__attach_campaign_sender_emails",
      "mcp__emailchaser__create_email_verification_job",
      "mcp__emailchaser__create_webhook",
      "mcp__emailchaser__detach_campaign_sender_email",
      "mcp__emailchaser__send_reply",
      "mcp__emailchaser__send_reply_draft",
      "mcp__emailchaser__set_billing_profile",
      "mcp__emailchaser__update_lead",
      "mcp__emailchaser__update_reply_draft",
      "mcp__emailchaser__update_sender_email",
      "mcp__emailchaser__update_sender_email_warmup",
      "mcp__emailchaser__update_webhook"
    ]
  }
}

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 18:37 UTC, unchanged since Nov 13, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
55
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
https://app.emailchaser.com/.well-known/oauth-protected-resource/api/mcp
Authorization servers
https://app.emailchaser.com
Legacy HTTP+SSE transport
no
Dynamic client registration
yes
Client ID metadata documents
no
  • HighThe whole server

    The endpoint served tools/list without credentials, and the list has 55 tools that change things (21 destructive, 1 irreversible, 33 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://app.emailchaser.com/api/mcp: tools/list returned 102 tools to an anonymous client, including launch_campaign (destructive), replace_campaign_sequence (destructive), delete_campaign (destructive), detach_campaign_sender_email (destructive), delete_lead (destructive) and 50 more

    What this check looks for

Posture changes

  1. Nov 13, 2026
    • RFC 9728 protected resource metadata: none to https://app.emailchaser.com/.well-known/oauth-protected-resource/api/mcp
    • Authorization servers: none to https://app.emailchaser.com
    • Dynamic client registration: not observed to yes
    • Client ID metadata documents: not observed to no

Findings

Matches of fixed checks, with the text they matched.

3 warn2 info
  • The server marks "get_outcomes_report" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"openWorldHint":false}; The name "get_outcomes_report" says "get": a word for reading.

    What this check looks for
  • The description contains "Every other tool", wording about other tools or servers.

    Matched text

    description: "...rks the workspace this key is bound to. Every other tool acts on the current workspace only; to..."

    What this check looks for
  • The server marks "cancel_email_verification_job" not destructive (destructiveHint: false), and its own name, description or input give the class "destructive".

    Matched text

    annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "cancel_email_verification_job" says "cancel": a word for removing or ending something.

    What this check looks for
  • The description is 2,273 characters long, over the 2,000 this check looks for.

    Matched text

    description ends: "...alse, and adding them is skipped for free too. A read-only key may use this tool."

    What this check looks for
  • The description is 2,681 characters long, over the 2,000 this check looks for.

    Matched text

    description ends: "...an importId, and the add's import record keeps changing until finishedAt is set."

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. PostureRemote posture changed. RFC 9728 protected resource metadata: none -> https://app.emailchaser.com/.well-known/oauth-protected-resource/api/mcp; Authorization servers: none -> https://app.emailchaser.com; Dynamic client registration: not obs…View diff
    • RFC 9728 protected resource metadata: none to https://app.emailchaser.com/.well-known/oauth-protected-resource/api/mcp
    • Authorization servers: none to https://app.emailchaser.com
    • Dynamic client registration: not observed to yes
    • Client ID metadata documents: not observed to no
  2. Tools102 changed.~102 changedView diff
  3. Tools102 changed.~102 changedView diff

Install

Pinned to version 1.0.1. Values in angle brackets are yours to fill in.

claude mcp add --transport http emailchaser https://app.emailchaser.com/api/mcp --header 'Authorization: <your-authorization>'
claude mcp add --transport http emailchaser https://app.emailchaser.com/api/mcp --header 'Authorization: <your-authorization>'

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__emailchaser__list_campaigns",
      "mcp__emailchaser__get_campaign",
      "mcp__emailchaser__get_campaign_sequence",
      "mcp__emailchaser__get_campaign_stats",
      "mcp__emailchaser__list_leads",
      "mcp__emailchaser__get_lead",
      "mcp__emailchaser__get_lead_conversation",
      "mcp__emailchaser__list_replies",
      "mcp__emailchaser__list_reply_drafts",
      "mcp__emailchaser__list_sender_emails",
      "mcp__emailchaser__get_sender_email",
      "mcp__emailchaser__get_sender_email_dns",
      "mcp__emailchaser__get_sender_email_warmup",
      "mcp__emailchaser__list_inbox_placement_tests",
      "mcp__emailchaser__list_inbox_placement_runs",
      "mcp__emailchaser__get_inbox_placement_run",
      "mcp__emailchaser__list_inbox_placement_run_results",
      "mcp__emailchaser__get_inbox_placement_stats_by_date",
      "mcp__emailchaser__get_sender_reputation",
      "mcp__emailchaser__get_deliverability_insights",
      "mcp__emailchaser__list_webhooks",
      "mcp__emailchaser__list_blocklist_entries",
      "mcp__emailchaser__get_blocklist_entry",
      "mcp__emailchaser__list_instantly_accounts",
      "mcp__emailchaser__list_icps",
      "mcp__emailchaser__get_primary_icp",
      "mcp__emailchaser__get_icp",
      "mcp__emailchaser__get_audience_size",
      "mcp__emailchaser__get_lead_finder_filters",
      "mcp__emailchaser__search_lead_finder",
      "mcp__emailchaser__get_lead_finder_search",
      "mcp__emailchaser__list_lead_finder_imports",
      "mcp__emailchaser__get_lead_finder_import",
      "mcp__emailchaser__list_autopilot_runs",
      "mcp__emailchaser__get_autopilot_run",
      "mcp__emailchaser__get_credit_balance",
      "mcp__emailchaser__list_credit_transactions",
      "mcp__emailchaser__search_dfy_domains",
      "mcp__emailchaser__list_dfy_orders",
      "mcp__emailchaser__get_dfy_order",
      "mcp__emailchaser__get_workspace_details",
      "mcp__emailchaser__list_workspaces",
      "mcp__emailchaser__get_billing_profile",
      "mcp__emailchaser__get_email_verification_rates",
      "mcp__emailchaser__list_email_verification_jobs",
      "mcp__emailchaser__get_email_verification_job",
      "mcp__emailchaser__list_email_verification_results"
    ],
    "ask": [
      "mcp__emailchaser__confirm_connection",
      "mcp__emailchaser__create_campaign",
      "mcp__emailchaser__update_campaign",
      "mcp__emailchaser__pause_campaign",
      "mcp__emailchaser__update_campaign_schedule",
      "mcp__emailchaser__attach_campaign_sender_emails",
      "mcp__emailchaser__move_lead_to_campaign",
      "mcp__emailchaser__add_leads",
      "mcp__emailchaser__update_lead",
      "mcp__emailchaser__mark_lead_meeting",
      "mcp__emailchaser__update_lead_category",
      "mcp__emailchaser__update_reply_draft",
      "mcp__emailchaser__connect_sender_email",
      "mcp__emailchaser__update_sender_email",
      "mcp__emailchaser__update_sender_email_warmup",
      "mcp__emailchaser__create_webhook",
      "mcp__emailchaser__update_webhook",
      "mcp__emailchaser__add_blocklist_entries",
      "mcp__emailchaser__update_blocklist_entry",
      "mcp__emailchaser__import_instantly",
      "mcp__emailchaser__copilot_plan",
      "mcp__emailchaser__copilot_launch",
      "mcp__emailchaser__create_icp",
      "mcp__emailchaser__update_icp",
      "mcp__emailchaser__set_primary_icp",
      "mcp__emailchaser__refresh_icp_audience_size",
      "mcp__emailchaser__plan_autopilot",
      "mcp__emailchaser__start_autopilot_run",
      "mcp__emailchaser__pause_autopilot_run",
      "mcp__emailchaser__resume_autopilot_run",
      "mcp__emailchaser__get_outcomes_report",
      "mcp__emailchaser__create_workspace",
      "mcp__emailchaser__create_workspace_api_key",
      "mcp__emailchaser__set_billing_profile"
    ],
    "deny": [
      "mcp__emailchaser__launch_campaign",
      "mcp__emailchaser__replace_campaign_sequence",
      "mcp__emailchaser__delete_campaign",
      "mcp__emailchaser__detach_campaign_sender_email",
      "mcp__emailchaser__delete_lead",
      "mcp__emailchaser__unmark_lead_meeting",
      "mcp__emailchaser__send_reply_draft",
      "mcp__emailchaser__send_reply",
      "mcp__emailchaser__delete_webhook",
      "mcp__emailchaser__remove_blocklist_entries",
      "mcp__emailchaser__remove_blocklist_entry",
      "mcp__emailchaser__delete_icp",
      "mcp__emailchaser__source_prospects",
      "mcp__emailchaser__add_lead_finder_prospects",
      "mcp__emailchaser__cancel_lead_finder_import",
      "mcp__emailchaser__approve_autopilot_run",
      "mcp__emailchaser__kill_autopilot_run",
      "mcp__emailchaser__purchase_credits",
      "mcp__emailchaser__create_dfy_order",
      "mcp__emailchaser__create_email_verification_job",
      "mcp__emailchaser__cancel_email_verification_job"
    ]
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as emailchaser. Review it before you use it.

Capability tags behind 58 rules
  • mcp__emailchaser__list_campaigns private-data, untrusted-input
  • mcp__emailchaser__get_campaign private-data, untrusted-input
  • mcp__emailchaser__launch_campaign untrusted-input
  • mcp__emailchaser__get_campaign_sequence private-data, untrusted-input
  • mcp__emailchaser__get_campaign_stats private-data
  • mcp__emailchaser__attach_campaign_sender_emails outbound
  • mcp__emailchaser__detach_campaign_sender_email outbound
  • mcp__emailchaser__get_lead private-data
  • mcp__emailchaser__update_lead outbound
  • mcp__emailchaser__get_lead_conversation private-data, untrusted-input
  • mcp__emailchaser__list_replies private-data, untrusted-input
  • mcp__emailchaser__list_reply_drafts untrusted-input
  • mcp__emailchaser__update_reply_draft outbound
  • mcp__emailchaser__send_reply_draft untrusted-input, outbound
  • mcp__emailchaser__send_reply untrusted-input, outbound
  • mcp__emailchaser__list_sender_emails private-data, untrusted-input
  • mcp__emailchaser__get_sender_email private-data, untrusted-input
  • mcp__emailchaser__connect_sender_email untrusted-input
  • mcp__emailchaser__update_sender_email outbound
  • mcp__emailchaser__get_sender_email_dns private-data, untrusted-input
  • mcp__emailchaser__get_sender_email_warmup private-data, untrusted-input
  • mcp__emailchaser__update_sender_email_warmup outbound
  • mcp__emailchaser__list_inbox_placement_tests private-data, untrusted-input
  • mcp__emailchaser__list_inbox_placement_runs private-data, untrusted-input
  • mcp__emailchaser__get_inbox_placement_run private-data, untrusted-input
  • mcp__emailchaser__list_inbox_placement_run_results private-data, untrusted-input
  • mcp__emailchaser__get_inbox_placement_stats_by_date private-data, untrusted-input
  • mcp__emailchaser__get_sender_reputation private-data, untrusted-input
  • mcp__emailchaser__create_webhook outbound
  • mcp__emailchaser__update_webhook outbound
  • mcp__emailchaser__list_blocklist_entries private-data, untrusted-input
  • mcp__emailchaser__get_blocklist_entry private-data, untrusted-input
  • mcp__emailchaser__import_instantly untrusted-input
  • mcp__emailchaser__list_instantly_accounts untrusted-input
  • mcp__emailchaser__copilot_plan untrusted-input
  • mcp__emailchaser__copilot_launch untrusted-input
  • mcp__emailchaser__refresh_icp_audience_size untrusted-input
  • mcp__emailchaser__get_audience_size untrusted-input
  • mcp__emailchaser__source_prospects untrusted-input
  • mcp__emailchaser__search_lead_finder private-data, untrusted-input
  • mcp__emailchaser__get_lead_finder_search untrusted-input
  • mcp__emailchaser__add_lead_finder_prospects untrusted-input
  • mcp__emailchaser__list_lead_finder_imports private-data, untrusted-input
  • mcp__emailchaser__get_lead_finder_import private-data, untrusted-input
  • mcp__emailchaser__start_autopilot_run untrusted-input
  • mcp__emailchaser__approve_autopilot_run untrusted-input
  • mcp__emailchaser__purchase_credits untrusted-input
  • mcp__emailchaser__get_outcomes_report private-data, untrusted-input
  • mcp__emailchaser__search_dfy_domains untrusted-input
  • mcp__emailchaser__create_dfy_order untrusted-input
  • mcp__emailchaser__list_dfy_orders private-data, untrusted-input
  • mcp__emailchaser__get_billing_profile private-data
  • mcp__emailchaser__set_billing_profile outbound
  • mcp__emailchaser__get_email_verification_rates private-data, untrusted-input
  • mcp__emailchaser__create_email_verification_job untrusted-input, outbound
  • mcp__emailchaser__list_email_verification_jobs private-data, untrusted-input
  • mcp__emailchaser__get_email_verification_job private-data, untrusted-input
  • mcp__emailchaser__list_email_verification_results private-data, untrusted-input

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/com.emailchaser/emailchaser/verify?hash=44934897f6b4e6c2a8d4d83a86f73e43b71bb0ccc5f33663148e18379635fcf8" | jq -e '.changed == false'

Pinned hash 44934897f6b4, from Nov 13, 2026, 03:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/com.emailchaser/emailchaser/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for com.emailchaser/emailchaser
[![Anlyon Radar](https://radar.anlyon.com/badge/com.emailchaser/emailchaser.svg)](https://radar.anlyon.com/servers/com.emailchaser/emailchaser)

Source: MCP Registry. Synced Oct 21, 2026, 17:54 UTC.

Emailchaser MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the Emailchaser MCP server safe?

No scan can promise that. What its definitions show: 102 tools, 55 of which can change, send or delete something, and 5 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the Emailchaser MCP server delete data?

Yes. 21 tools can delete, cancel or revoke something: launch_campaign, replace_campaign_sequence, delete_campaign, detach_campaign_sender_email, delete_lead, unmark_lead_meeting and 15 more.

Can the Emailchaser MCP server do things that cannot be undone?

Yes. 1 tool can send, pay, publish or merge: update_reply_draft.

What tools does the Emailchaser MCP server have?

102 tools: 21 destructive, 1 irreversible, 33 write, 47 read. launch_campaign, replace_campaign_sequence, delete_campaign, detach_campaign_sender_email, delete_lead, unmark_lead_meeting, send_reply_draft, send_reply and 94 more.

Does the Emailchaser MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the Emailchaser MCP server send data to outside services?

12 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): attach_campaign_sender_emails, detach_campaign_sender_email, update_lead, update_reply_draft, send_reply_draft, send_reply and 6 more.

Has the Emailchaser MCP server changed its tools?

Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 13, 2026. Every change is on this page and in its Atom feed.