Skip to content

Jid: the trust API for AI agents MCP server

com.jid/jidjid.comstreamable-httpv1.6.0Scanned

Verify a business before an agent pays, signs or buys: registry-cited decision and signed receipt.

7 tools: 1 do something that can't be undone, 3 change things and 3 only read.

Tools: 0 delete, 1 irreversible, 3 write, 3 read.
  • Reads private data · 2
  • Reads text others write · 3
  • Can send data out · 1
  • 1 finding

Tools listed without sign-in. Remote posture

With Anlyon

1 tool here does something that can't be undone. Govern it.

  • 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 Jid: the trust API for AI agents MCP server can do

The Jid: the trust API for AI agents MCP server is published by jid.com, a domain the MCP Registry verified. It has 7 tools. It can do things that cannot be undone, such as sending, paying or publishing (verify_business) and create or change data (3 tools: qualify_counterparty, screen and check_email), but cannot delete data. 3 tools only read.

2 tools read private data (search_business and get_business_file); 3 tools bring in text other people can write (qualify_counterparty, screen and check_email); 1 tool can send data out of your system (screen). 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 1 finding (annotation-conflict). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in.

Last scanned Nov 13, 2026, version 1.6.0.

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.

7 tools

Can't be undone 1Claude Code: ask
  • verify_businessVerifies a business against the registry facts Jid holds.1 finding

    Why it is in this group

    • The input carries an amount (input.amount) and a currency, so it moves money.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    Findings

    • Warn

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

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":false,"openWorldHint":false}; The input carries an amount (input.amount) and a currency, so it moves money.

      What this check looks for
    Description, as the server wrote it
    Verifies a business against the registry facts Jid holds. Input: a Jid id (or a name or registry number) and, optionally, an activity, jurisdiction and amount. With an activity it returns a decision (proceed, review, stop or insufficient_data) with reasons: each names the requirement, met (true, false or unknown), the fact ids, source, verified_at and fresh_until, plus a signed receipt_id. Without an activity it returns the status and the top cited facts. Default output is compact (verdict, reasons, status, identifiers, top facts, receipt reference); detail "full" adds every cited fact and the full signed receipt. Activities: general_commerce (General commerce (default)); sell_prescription_drugs (Sell prescription drugs); manufacture_drugs (Manufacture drugs); sell_medical_devices (Sell medical devices); haul_freight (Haul freight (for-hire motor carrier)); broker_freight (Broker freight); sell_alcohol (Sell alcohol (wholesale or import)); produce_spirits (Produce distilled spirits); provide_investment_advice (Provide investment advice); accept_deposits (Accept deposits); lend (Lend); bill_medicare_medicaid (Bill Medicare or Medicaid); healthcare_supplier (Health care supplier); federal_contracting (Federal contracting); charitable_solicitation (Charitable solicitation); construction (Construction or home improvement (contractor)); operate_psilocybin_healing_center (Operate a psilocybin healing or service centre (CO, OR)).

    Annotations the server declares

    • title: Verify a business before acting
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: false
Changes things 3Claude Code: ask
  • check_emailChecks whether an e-mail is from who it claims.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
    Checks whether an e-mail is from who it claims. Input: the raw headers of a message (as shown by Show original or View source). Returns checks and a read (headline, recommendation engage | verify_first | do_not_engage, failed checks, lookup links): DMARC/SPF/DKIM as the receiving server recorded them, DKIM alignment with From, Reply-To going to another domain, and a From domain or display name that imitates a known business. The headers are processed in memory and never stored. Deterministic, no AI. Counts as 1 check with a key; free tier without one.

    Annotations the server declares

    • title: Is this e-mail really from who it says? (SPF, DKIM, DMARC, lookalike sender)
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
  • qualify_counterpartyQualifies a business that made contact (a form submission, an inquiry, an outreach reply).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
    Qualifies a business that made contact (a form submission, an inquiry, an outreach reply). Input: the sender's e-mail (or the company domain), the company name if given, and the context (form type, message, stated value, stated role). Jid checks the domain (MX, SPF, DMARC, registration date over RDAP, the website), matches it to a Jid business file, reads that file's registrations, licences, filings, capital raises, loans and adverse records, and returns signals (each with source, source_url, verified_at), three scores (legitimacy, capacity, consistency, 0 to 100 with reasons), checks (each pass, fail or unknown with one line of evidence) and a read: a one-line headline ("Passed all 7 checks Jid could run. Recommendation: engage."), the recommendation (engage, verify_first, do_not_engage), the failed checks and up to 4 links to verify by hand, plus a signed receipt reference. detail "full" adds each signal's raw value, each check's lookup link and the full signed receipt. Businesses only: personal webmail or no matched business returns individual_or_unverified with no scores. A possible list match is returned as 'review', not as a finding. Counts as 2 checks with a key; free tier without one. Not intended for credit, employment, insurance or housing decisions about any person.

    Annotations the server declares

    • title: Qualify a business lead or counterparty
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
  • screenScreens a lead, sender, vendor or website and returns an engage-or-not answer.Reads text others writeCan send data outNo 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.
    • outboundThe input takes a destination (input.phone): what the tool writes goes there.
    • 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
    Screens a lead, sender, vendor or website and returns an engage-or-not answer. Input: any of domain, url, email, phone or company_name, the context (lead, inbound_email, partnership, vendor, purchase, marketplace_seller, agent_action), an optional amount (USD) and a policy (strict, balanced, permissive; default balanced). Returns engage (true or false), risk (low, medium, high, unknown), confidence 0 to 1, subject_type, reasons (each with source and evidence), mitigations, checks, a read (headline, recommendation engage | verify_first | do_not_engage, failed checks, lookup links) and a signed receipt_id. Deterministic, no AI in the decision, usually under a second. A possible sanctions name match returns engage=false with 'manual review', not a finding. For individuals only technical checks of the address, phone or domain run. engage is advisory. Rules: https://jid.com/screen-rules. Counts as 1 check with a key; free tier without one.

    Annotations the server declares

    • title: Should I engage? (screen a lead, sender, vendor or website)
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
Only reads 3Show 3 read-only tools
  • get_business_fileThe Jid file for one business.Reads private dataNo findings

    Why it is in this group

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

    What it can reach

    • private-dataThe name "get_business_file" says "get" and "file": it reads files.
    • private-dataThe description's first sentence says "file": it reads files.
    Description, as the server wrote it
    The Jid file for one business. Default (compact): status with its flags, identifiers, verified websites, the top facts with source and verification date, counts and the latest changes. With detail "full": every current fact with its source, verification date, next check and confidence, owner statements (labelled self-reported), possible matches and the change history.

    Annotations the server declares

    • title: Get a business file
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • get_receiptReturns a stored, Ed25519-signed verification receipt by receipt_id: the inputs, decision, reasons, fact ids and value hashes, rules version and time of the check.No findings

    Why it is in this group

    • The name "get_receipt" says "get": a word for reading.
    Description, as the server wrote it
    Returns a stored, Ed25519-signed verification receipt by receipt_id: the inputs, decision, reasons, fact ids and value hashes, rules version and time of the check. Free; does not count as a check.

    Annotations the server declares

    • title: Get a verification receipt
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • search_businessFinds Jid business files by name or by any registry number (FEI, LEI, SEC CIK, DUNS, NDC labeler code, state licence or entity number, French SIREN or SIRET).Reads private dataNo findings

    Why it is in this group

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

    What it can reach

    • private-dataThe description's first sentence says "find" and "file": it reads files.
    Description, as the server wrote it
    Finds Jid business files by name or by any registry number (FEI, LEI, SEC CIK, DUNS, NDC labeler code, state licence or entity number, French SIREN or SIRET). Returns candidate files with status and match evidence. For a French company pass country "FR": the SIRENE register is searched live and files are created on demand.

    Annotations the server declares

    • title: Search businesses
    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false

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.

6 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 outbound tool breaks every one.

Reads private data2
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write3
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out1
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__jid__screen"
    ]
  }
}

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, 18:40 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
4
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 4 tools that change things (1 irreversible, 3 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://jid.com/mcp: tools/list returned 7 tools to an anonymous client, including verify_business (irreversible), qualify_counterparty (write), screen (write), check_email (write)

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

1 warn
  • The server marks "verify_business" read-only (readOnlyHint: true), and its own name, description or input give the class "irreversible".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":false,"openWorldHint":false}; The input carries an amount (input.amount) and a currency, so it moves money.

    What this check looks for

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 1.6.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http jid https://jid.com/mcp
claude mcp add --transport http jid https://jid.com/mcp

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__jid__search_business",
      "mcp__jid__get_business_file",
      "mcp__jid__get_receipt"
    ],
    "ask": [
      "mcp__jid__verify_business",
      "mcp__jid__qualify_counterparty",
      "mcp__jid__screen",
      "mcp__jid__check_email"
    ],
    "deny": []
  }
}

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 jid. Review it before you use it.

Capability tags behind 5 rules
  • mcp__jid__search_business private-data
  • mcp__jid__get_business_file private-data
  • mcp__jid__qualify_counterparty untrusted-input
  • mcp__jid__screen untrusted-input, outbound
  • mcp__jid__check_email 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.jid/jid/verify?hash=a2dfe95c7ace80ab056f3907fc568af5b5c22d16fe4493b01151bf854818e797" | jq -e '.changed == false'

Pinned hash a2dfe95c7ace, from Nov 13, 2026, 13:18 UTC.

Watch

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

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

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

Badge

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

Source: MCP Registry. Synced Oct 21, 2026, 17:57 UTC.

Jid: the trust API for AI agents MCP server: questions and answers

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

Is the Jid: the trust API for AI agents MCP server safe?

No scan can promise that. What its definitions show: 7 tools, 4 of which can change, send or delete something, and 1 finding from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the Jid: the trust API for AI agents MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the Jid: the trust API for AI agents MCP server do things that cannot be undone?

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

What tools does the Jid: the trust API for AI agents MCP server have?

7 tools: 1 irreversible, 3 write, 3 read. verify_business, qualify_counterparty, screen, check_email, search_business, get_business_file and get_receipt.

Does the Jid: the trust API for AI agents MCP server need sign-in or OAuth?

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

Does the Jid: the trust API for AI agents MCP server send data to outside services?

1 tool can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): screen.

Has the Jid: the trust API for AI agents MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 13, 2026.