PostAgent MCP server
tech.interpretai/PostAgentinterpretai.techstreamable-httpv0.5.0Scanned
Print & mail PDF/HTML/Markdown/text/DOCX/images to US addresses; pay per call in x402 USDC on Base.
14 tools: 2 can delete or cancel things, 5 do something that can't be undone, 5 change things and 2 only read.
- Reads private data · 1
- Reads text others write · 14
- Can send data out · 8
- 3 findings
Tools listed without sign-in. Remote posture
With Anlyon
2 tools here can delete or cancel things and 5 do 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.
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 PostAgent MCP server can do
The PostAgent MCP server is published by interpretai.tech, a domain the MCP Registry verified. It has 14 tools. It can delete or cancel things (2 tools: pay_mail_with_shared_payment_token and submit_paid_mail_job), do things that cannot be undone, such as sending, paying or publishing (check_postagent_updates, create_mail_quote, prepare_mail_payment, create_postcard_quote and 1 more) and create or change data (5 tools: create_letter, create_template, create_card_checkout, create_postcard_art and 1 more). 2 tools only read.
1 tool reads private data (get_mail_job_status); 14 tools bring in text other people can write (check_postagent_updates, create_letter, create_template and 11 more); 8 tools can send data out of your system (create_letter, create_template, create_mail_quote and 5 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 3 findings (annotation-conflict and credentials-mention). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 18, 2026, version 0.5.0.
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.
14 tools
Deletes or cancels 2Claude Code: deny
submit_paid_mail_jobDetached-signature fallback for x402 wallets that can emit a standalone PAYMENT-SIGNATURE header.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "submit_paid_mail_job" says "submit": a word for a change.
- The description says "IRREVERSIBLE".
- 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 "submit_paid_mail_job" says "submit" and "mail": it writes something other people read.
Description, as the server wrote it Detached-signature fallback for x402 wallets that can emit a standalone PAYMENT-SIGNATURE header. THE PRIMARY/RECOMMENDED PATH is for the agent's wallet to pay the quote's `paymentUrl` in-band (e.g. `npx awal@latest x402 pay <paymentUrl>`); use this tool only if your wallet client cannot do that. Charges the agent in USDC on Base mainnet and creates a physical letter for printing and mailing. THIS IS IRREVERSIBLE. Only call after the user has explicitly confirmed the recipient, sender, content, and price returned by create_mail_quote, and after obtaining the signed x402 payment header (see prepare_mail_payment).
Annotations the server declares
- readOnlyHint: false
- destructiveHint: true
- idempotentHint: false
- openWorldHint: true
Can't be undone 5Claude Code: ask
check_postagent_updatesFetches the live PostAgent agent manifest.Reads text others write1 finding
Why it is in this group
- The name "check_postagent_updates" says "update": a word for a change.
- The description says "irreversible".
- 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.
Findings
- Warn
The server marks "check_postagent_updates" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "check_postagent_updates" says "update": a word for a change.
Description, as the server wrote it Fetches the live PostAgent agent manifest. Call this before using PostAgent in a new session, after reconnecting the MCP server, or when an installed PostAgent skill may be stale. If the installed skill is older than latestSkillVersion, read latestSkillUrl and follow those instructions for this turn; if updateRequired is true, do not perform paid or irreversible PostAgent actions until the user updates.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
create_campaign_quoteQuote ONE bulk send to MANY recipients (up to 500): a single template or static document, fulfilled through PostAgent's campaign workflow.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "create_campaign_quote" says "create": a word for a change.
- The input addresses people (input.recipients): what it sends cannot be taken back.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a destination (input.recipients): what the tool writes goes there.
Description, as the server wrote it Quote ONE bulk send to MANY recipients (up to 500): a single template or static document, fulfilled through PostAgent's campaign workflow. Price is an inclusive customer-facing total, locked for 15 minutes. Payment is x402 ONLY. IMPORTANT — KYC WALL: campaigns require the paying wallet to be identity-verified with PostAgent. Verification is operator-run, not self-serve: until the wallet is verified this tool returns `kyc_required` (403). Do not retry the same wallet — instead either mail recipients individually with create_mail_quote, or request campaign access at https://interpretai.tech/contact (this URL is also returned in the 403's details.contactUrl) to get the wallet enabled. Usage once the wallet is verified: upload a template with create_template (using {{fields}}) or a finished letter with create_letter, then call this with `recipients` (each `to` a US address, plus per-recipient `mergeVariables` for templates) and the `payerWallet` that will pay the quote. The print partner validates every recipient address after payment; failed recipients are excluded and their share refunded (processed manually). Track via get_campaign_status.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
create_mail_quoteVerifies the recipient and sender US addresses and locks a 15-minute USDC price for a documentId.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "create_mail_quote" says "create": a word for a change.
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe name "create_mail_quote" says "create" and "mail": it writes something other people read.
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it Verifies the recipient and sender US addresses and locks a 15-minute USDC price for a documentId. Does not charge or mail anything. Returns a `paymentUrl` (a per-quote x402-payable URL); the preferred way to actually mail the letter is for the agent's wallet to perform an in-band x402 payment against that URL (e.g. `npx awal@latest x402 pay <paymentUrl>`). The MCP `submit_paid_mail_job` tool is a fallback for clients that can emit a standalone signature header. In all cases, show the recipient, sender, options, selected-route `design` constraints, price, AND any `fulfillment.warnings` to the user and get explicit confirmation before paying. The response includes a `fulfillment` block with `requested` (what you asked for), `selected` (what will actually be printed/mailed) and `warnings` (any soft-preference downgrades — e.g. `service_level_downgraded` or `extra_service_unavailable`); do not pay through a non-empty warnings list without re-confirming the trade-off with the user. The response also includes a provider-neutral `design` block; inspect it and the preview before paying because the selected delivery method determines print address/no-ink zones.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
create_postcard_quoteVerifies the recipient and sender addresses and locks a 15-minute USDC price for mailing a postcard built from two postcard_art documents (front + back, uploaded via create_postcar…Reads text others writeCan send data outNo findings
Why it is in this group
- The name "create_postcard_quote" says "create": a word for a change.
- The input addresses people (input.to): what it sends cannot be taken back.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- outboundThe input takes a destination (input.to): what the tool writes goes there.
Description, as the server wrote it Verifies the recipient and sender addresses and locks a 15-minute USDC price for mailing a postcard built from two postcard_art documents (front + back, uploaded via create_postcard_art). Sizes: 4x6 (default), 6x9, 6x11 — postcards always print in full color. US recipients use the strict US address shape; INTERNATIONAL recipients are supported on 4x6 only (set `to.country` to the 2-letter ISO code; the sender must still be a US address). Payment works exactly like letters: pay the returned `paymentUrl` via x402, or use the MPP/checkout fallbacks. Show the recipient, sender, size, selected-route `design` constraints, price, AND any `fulfillment.warnings` to the user and get explicit confirmation before paying — service level is a soft preference and may have been downgraded (e.g. international economy is upgraded to standard) with a `service_level_downgraded` warning the agent must surface verbatim.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
prepare_mail_paymentReturns the x402 PAYMENT-REQUIRED challenge for a locked quote so an x402-capable wallet client can sign it.Reads text others write1 finding
Why it is in this group
- The name "prepare_mail_payment" says "payment": a word that usually means money moves, which cannot be taken back.
- 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.
Findings
- Warn
The server marks "prepare_mail_payment" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "prepare_mail_payment" says "payment": a word that usually means money moves, which cannot be taken back.
Description, as the server wrote it Returns the x402 PAYMENT-REQUIRED challenge for a locked quote so an x402-capable wallet client can sign it. No payment is taken at this step. Probes the canonical per-quote pay URL (`/v1/quotes/:quoteId/pay`). The preferred way to actually pay is for the wallet to perform the standard x402 in-band handshake against `paymentUrl`; this tool is for inspection or for the detached-signature flow via `submit_paid_mail_job`.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
Changes things 5Claude Code: ask
create_card_checkoutCreates a Stripe-hosted Checkout page for a locked quote and returns a `checkoutUrl` plus a `statusUrl`.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "create_card_checkout" says "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.
- outboundThe input takes a URL (input.webhookUrl): a request to a URL can carry data to whoever runs it.
Description, as the server wrote it Creates a Stripe-hosted Checkout page for a locked quote and returns a `checkoutUrl` plus a `statusUrl`. Use this as the LAST-RESORT payment rail, when the payer is a human paying by credit card (the agent cannot complete card entry itself). Hand the `checkoutUrl` to the user to open in a browser and pay; payment is asynchronous — once they finish, the letter is created automatically by Stripe's webhook. To track it without a job id, GET the returned `statusUrl` (`/v1/quotes/:quoteId/job`): it returns `found:false` with status `pending`/`processing` until the webhook creates the job, then the full job (id, status, tracking). Prefer x402 (or MPP, if available) for autonomous agent payment. Returns an error if Stripe Checkout is not enabled on the server. No charge occurs until the human completes the hosted page.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
create_letterUpload and normalize a FINISHED, ready-to-mail document to PDF.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "create_letter" says "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.
- 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 Upload and normalize a FINISHED, ready-to-mail document to PDF. Choose this when the content is final and IDENTICAL for every recipient — including when you mail the same letter to many people (just quote/pay once per recipient with the same documentId). The exact bytes you give are what gets printed. Use create_template instead only when the content must vary per recipient via {{fields}}. Returns a documentId, the stored page count, byte size, and source format. Free; no payment required. Provide the document EXACTLY ONE way: `content` (inline text, for html/markdown/text), `contentBase64` (base64-encoded binary, for pdf/docx/image), or `url` (a publicly reachable URL the server fetches). Supplying none, or more than one, is an error. Maximum upload size is 31457280 bytes (~30 MB); output page size is US Letter. Any `{{...}}` text is printed LITERALLY here — it is NOT treated as a merge field. If you want personalized mail merge across recipients, use `create_template` instead. Reserved address zone: a recipient address block is printed over the top ~3 inches of page 1, so the server reserves that space for you automatically. For text/html/markdown/docx, page-1 content is pushed below the block (content may therefore flow onto an additional page); for pdf and image inputs, a blank first page is prepended. As a result the returned page count — and the selected-provider cost behind the resulting quote — can be higher than your source document (e.g. a single-page PDF is stored as 2 pages). You do NOT need to leave the top of your document blank yourself. See the postagent://formats resource for per-format details.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
create_postcard_artUpload ONE side of a postcard (front or back) as single-page PDF, PNG, or JPEG artwork.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "create_postcard_art" says "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.
- 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 Upload ONE side of a postcard (front or back) as single-page PDF, PNG, or JPEG artwork. The file is stored VERBATIM — no page-size normalization and no reserved address zone — so you are responsible for the correct trim size plus bleed: 4x6 cards need 4.25"x6.25" artwork, 6x9 needs 6.25"x9.25", 6x11 needs 6.25"x11.25" (0.125" bleed on each edge). The selected delivery method prints the recipient address block over part of the BACK, so keep that area clear of critical content and review the quote's `design` block before payment. Returns a documentId with kind "postcard_art". Upload front and back separately, then quote with create_postcard_quote. Free; no payment required. Provide the artwork EXACTLY ONE way: `contentBase64` (base64-encoded pdf/png/jpeg) or `url` (publicly reachable). Text-based formats are rejected — postcards are artwork, not documents.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
create_templateUpload a REUSABLE template containing `{{field}}` placeholders (e.g.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "create_template" says "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.
- 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 Upload a REUSABLE template containing `{{field}}` placeholders (e.g. `Dear {{name}},` or `Balance due: {{amount}}`). Choose this ONLY when the content must vary per recipient (mail merge) — recipient count is irrelevant, so a single personalized letter belongs here too. If the content is identical for everyone, use create_letter instead (this tool rejects input with no `{{fields}}`). Returns a documentId with `kind: "html_template"`, a `mergeFields` list of the detected field names, and an `estimatedPageCount`. Free; no payment required. Template source must be TEXT-BASED (html, markdown, or text) and must contain at least one `{{field}}`, or the upload is rejected — for a finished document with no merge fields, use `create_letter`. Provide the template EXACTLY ONE way: `content` (inline text), `contentBase64` (base64-encoded text), or `url` (a publicly reachable URL the server fetches). Supplying none, or more than one, is an error. Maximum upload size is 31457280 bytes (~30 MB); output page size is US Letter. Reuse one template documentId across recipients: call create_mail_quote ONCE PER RECIPIENT, supplying that recipient's values via `mergeVariables` (every field in `mergeFields` must have a non-empty value). The server substitutes the values and renders that recipient's personalized PDF at quote time, so `estimatedPageCount` is only a baseline — the binding page count and price are set per quote from the actual rendered output. Reserved address zone: a recipient address block is printed over the top ~3 inches of page 1, so the server reserves that space automatically (page-1 content is pushed below the block and may flow onto an additional page). You do NOT need to leave the top blank yourself. See the postagent://formats resource for details.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
verify_addressStandalone paid address verification — no mail is sent.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 Standalone paid address verification — no mail is sent. Checks whether an address is deliverable and returns the standardized form (US: CASS with ZIP+4; international: per-country matching). Costs a small flat USDC fee per call via x402 (a fraction of a cent vs. mailing). Two-step flow, like the mail rails: call WITHOUT `paymentSignature` to get the 402 challenge and `paymentUrl` (preferred: have the agent's x402 wallet pay `paymentUrl` in-band with the address as the JSON POST body `{"address":{...}}`); or sign the challenge and call again WITH `paymentSignature` to verify and get the result in one round trip. US addresses need line1 + (city+state or zip). International addresses need line1 + country. Note: when mailing through PostAgent you do NOT need this tool — create_mail_quote already verifies sender and recipient for free as part of the quote.
Annotations the server declares
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
Only reads 2Show 2 read-only tools
get_campaign_statusReturns a campaign's validation/sending progress: status (validating | sent | partial | failed), recipient counts (total / validated / failed), and a failures report URL when some …Reads text others writeNo findings
Why it is in this group
- The name "get_campaign_status" 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.
- untrusted-inputThe description's first sentence says "return" and "url": it reads web pages, which anyone can write.
Description, as the server wrote it Returns a campaign's validation/sending progress: status (validating | sent | partial | failed), recipient counts (total / validated / failed), and a failures report URL when some recipients could not be validated. Polling this endpoint also advances the campaign (it releases the campaign for mailing once the print partner finishes validating the audience). Find the campaign id on the quote's job status (GET jobStatusUrl) after payment.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
get_mail_job_statusReturns normalized status, carrier/tracking/proof data, and tracking events for a previously created job.Reads private dataReads text others writeNo findings
Why it is in this group
- The name "get_mail_job_status" says "get": a word for reading.
What it can reach
- private-dataThe name "get_mail_job_status" says "get" and "mail": it reads mail.
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
- untrusted-inputThe name "get_mail_job_status" says "get" and "mail": it reads inbound email, which other people can write.
Description, as the server wrote it Returns normalized status, carrier/tracking/proof data, and tracking events for a previously created job.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- 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.
112 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 private-data tool breaks every one.
- Reads private data1
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write14
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out8
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
{
"permissions": {
"deny": [
"mcp__postagent__get_mail_job_status"
]
}
}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.
Looked at Nov 20, 2026, 19:19 UTC, unchanged since Oct 22, 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
- 12
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
- HighThe whole server
The endpoint served tools/list without credentials, and the list has 12 tools that change things (2 destructive, 5 irreversible, 5 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://api.postagent.sh/mcp: tools/list returned 14 tools to an anonymous client, including pay_mail_with_shared_payment_token (destructive), submit_paid_mail_job (destructive), check_postagent_updates (irreversible), create_mail_quote (irreversible), prepare_mail_payment (irreversible) and 7 more
Findings
Matches of fixed checks, with the text they matched.
The server marks "check_postagent_updates" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "check_postagent_updates" says "update": a word for a change.
The server marks "prepare_mail_payment" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".
What this check looks forMatched text
annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The name "prepare_mail_payment" says "payment": a word that usually means money moves, which cannot be taken back.
The description mentions "credential".
What this check looks forMatched text
description: "...eate … --network-id <stripeProfileId> --credential-type shared_payment_token`) or the Shar..."
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 0.5.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http postagent https://api.postagent.sh/mcpclaude mcp add --transport http postagent https://api.postagent.sh/mcpRun 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.
{
"permissions": {
"allow": [
"mcp__postagent__get_mail_job_status",
"mcp__postagent__get_campaign_status"
],
"ask": [
"mcp__postagent__check_postagent_updates",
"mcp__postagent__create_letter",
"mcp__postagent__create_template",
"mcp__postagent__create_mail_quote",
"mcp__postagent__create_card_checkout",
"mcp__postagent__prepare_mail_payment",
"mcp__postagent__create_postcard_art",
"mcp__postagent__create_postcard_quote",
"mcp__postagent__verify_address",
"mcp__postagent__create_campaign_quote"
],
"deny": [
"mcp__postagent__pay_mail_with_shared_payment_token",
"mcp__postagent__submit_paid_mail_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 postagent. Review it before you use it.
Capability tags behind 14 rules
- mcp__postagent__check_postagent_updates untrusted-input
- mcp__postagent__create_letter untrusted-input, outbound
- mcp__postagent__create_template untrusted-input, outbound
- mcp__postagent__create_mail_quote untrusted-input, outbound
- mcp__postagent__create_card_checkout untrusted-input, outbound
- mcp__postagent__pay_mail_with_shared_payment_token untrusted-input
- mcp__postagent__prepare_mail_payment untrusted-input
- mcp__postagent__submit_paid_mail_job untrusted-input, outbound
- mcp__postagent__get_mail_job_status private-data, untrusted-input
- mcp__postagent__create_postcard_art untrusted-input, outbound
- mcp__postagent__create_postcard_quote untrusted-input, outbound
- mcp__postagent__verify_address untrusted-input
- mcp__postagent__create_campaign_quote untrusted-input, outbound
- mcp__postagent__get_campaign_status 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/tech.interpretai/PostAgent/verify?hash=b94110f5aa2e0b0599189bd5fe79869603692867a83af8321dddc8219d240f6b" | jq -e '.changed == false'Pinned hash b94110f5aa2e, from Nov 18, 2026, 08:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/tech.interpretai/PostAgent/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/tech.interpretai/PostAgent)Source: MCP Registry · repository. Synced Oct 21, 2026, 18:55 UTC.
PostAgent MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the PostAgent MCP server safe?
- No scan can promise that. What its definitions show: 14 tools, 12 of which can change, send or delete something, and 3 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 PostAgent MCP server delete data?
- Yes. 2 tools can delete, cancel or revoke something: pay_mail_with_shared_payment_token and submit_paid_mail_job.
Can the PostAgent MCP server do things that cannot be undone?
- Yes. 5 tools can send, pay, publish or merge: check_postagent_updates, create_mail_quote, prepare_mail_payment, create_postcard_quote and create_campaign_quote.
What tools does the PostAgent MCP server have?
- 14 tools: 2 destructive, 5 irreversible, 5 write, 2 read. pay_mail_with_shared_payment_token, submit_paid_mail_job, check_postagent_updates, create_mail_quote, prepare_mail_payment, create_postcard_quote, create_campaign_quote, create_letter and 6 more.
Does the PostAgent MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the PostAgent MCP server send data to outside services?
- 8 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): create_letter, create_template, create_mail_quote, create_card_checkout, submit_paid_mail_job, create_postcard_art and 2 more.
Has the PostAgent MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 18, 2026.