Skip to content

Rightshold MCP server

com.rightshold/rightsholdrightshold.comstreamable-httpv0.5.1Scanned

License checks, signed rights proof, keyless pre-call checks for MCP servers, and the agent store

24 tools: 19 change things and 5 only read.

Tools: 0 delete, 0 irreversible, 19 write, 5 read.
  • Reads private data · 1
  • Reads text others write · 7
  • Can send data out · 1
  • 10 findings

Tools listed without sign-in. Remote posture

With Anlyon

19 tools here change things. 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 Rightshold MCP server can do

The Rightshold MCP server is published by rightshold.com, a domain the MCP Registry verified. It has 24 tools. It can create or change data (19 tools: start_here, read_ai_policy, license_compatibility, attribution_string and 15 more), but cannot delete data. 5 tools only read.

1 tool reads private data (kya_lookup); 7 tools bring in text other people can write (read_ai_policy, check_license, check_rights_batch and 4 more); 1 tool can send data out of your system (check_listing). 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 10 findings (annotation-conflict and credentials-mention). 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 14, 2026, version 0.5.1.

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.

24 tools

Changes things 19Claude Code: ask
  • ask_rightsholdCall when you have a rights or licensing question in plain English, such as 'Can I use CC BY-NC 4.0 images in a paid ad campaign?'.No findings

    Why it is in this group

    • The description's first verb is "call": a word for a change.
    Description, as the server wrote it
    Call when you have a rights or licensing question in plain English, such as 'Can I use CC BY-NC 4.0 images in a paid ad campaign?'. Returns the license or registered resource and the use it understood, and a deterministic answer with reasons; it asks for clarification instead of guessing. Check the understood use before relying on it. Free, no key.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: false
  • assess_resource_for_useCall when the asset is a registered resource and you need its exact answer for one use.No findings

    Why it is in this group

    • The description's first verb is "call": a word for a change.
    Description, as the server wrote it
    Call when the asset is a registered resource and you need its exact answer for one use. Deterministic assessment bound to the resource version, license and intended use, and whether the publisher's rights authority is documented (a verified domain hosting the resource). Four dispositions with reason codes; no trust score. Free, no key.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: false
  • attribution_stringCall when you reuse material that requires credit.1 finding

    Why it is in this group

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

    Findings

    • Warn

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

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "call": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Call when you reuse material that requires credit. Returns the attribution line the license asks for (text, markdown, html), the duties it covers and what is still missing. Free, no key. Call with {"license": "CC-BY-4.0", "title": "Sunset", "author": "A. Photographer", "source_url": "https://example.org/sunset"}.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • check_licenseCall before using any image, text, dataset, code, font or model you did not create.Reads text others write1 finding

    Why it is in this group

    • The description's first verb is "call": a word for a change.
    • 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_license" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The description's first verb is "call": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Call before using any image, text, dataset, code, font or model you did not create. Checks one license (any SPDX id; curated readings for common ones, needs_review for the rest) or the license a GitHub, Hugging Face, npm, PyPI, Wikimedia, Unsplash or Pexels page declares (source_url), for your intended use. Returns permitted, prohibited, needs_review or not_documented, with duties and an uncertainty block (what the verdict leaves open; whether a human must review). Free, no key. Call with {"license": "CC-BY-4.0", "intended_use": {"commercial_use": true, "attribution_planned": true}}.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
    • untrustedContentHint: true
  • check_listingPre-install check against the Rightshold Store and its signed revocation list: give a package hash (SHA-256), an endpoint or package URL, or a store slug.Can 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

    • 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
    Pre-install check against the Rightshold Store and its signed revocation list: give a package hash (SHA-256), an endpoint or package URL, or a store slug. Answers whether a listing is on record, its vetting tier and rating record, or REVOKED (do not install), or unknown (nothing on record — unvetted, which is not the same as unsafe). Call with {"slug": "rightshold-rights"}.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • check_rights_batchCall before shipping anything that bundles third-party assets.Reads text others writeNo findings

    Why it is in this group

    • The description's first verb is "call": 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
    Call before shipping anything that bundles third-party assets. Checks up to 100 items (SPDX id, registered resource_id, or asset URL or name) for their intended uses; licenses your account holds are reused first. Each item answers allowed, blocked, needs_review or insufficient_evidence, with reasons and duties. Free, no key. Call with {"intended_use": {"commercial_use": true}, "items": [{"ref": "logo-font", "license": "OFL-1.1"}, {"ref": "hero-photo", "license": "CC-BY-NC-4.0"}]}.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
  • gate_checkCall when your site accepts agents or uploads.Reads text others write1 finding

    Why it is in this group

    • The description's first verb is "call": a word for a change.
    • 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 "gate_check" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The description's first verb is "call": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Call when your site accepts agents or uploads. Free, no key. The Gate websites use on agent traffic: checks an RS-KYA token (or handle), a Web Bot Auth signature from any resolvable Signature-Agent signer, a Rights Receipt, clearance and intent token against the site's policy (min_kya_level, require_kya, require_web_bot_auth, allowed_signature_agents, require_receipt, allow_agents, deny_agents, require_clearance, require_intent, max_intent_risk); allow or deny, with reasons.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
    • untrustedContentHint: true
  • issue_rights_manifestUse when your principal wants signed proof for a whole deliverable: checks every third-party input (resource_id, license or asset) for its use, reusing licenses you hold, and retur…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
    Use when your principal wants signed proof for a whole deliverable: checks every third-party input (resource_id, license or asset) for its use, reusing licenses you hold, and returns a signed, ledger-committed manifest plus a text summary; verify_rights_manifest checks it. Call with {"title": "Spring campaign", "intended_use": {"commercial_use": true}, "items": [{"ref": "font", "license": "OFL-1.1"}]}.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: false
  • issue_rights_receiptUse when your principal wants signed proof that one asset you ship was cleared: a Rights Receipt (open standard rights-receipt/1) binding the file's SHA-256 to its license, intende…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
    Use when your principal wants signed proof that one asset you ship was cleared: a Rights Receipt (open standard rights-receipt/1) binding the file's SHA-256 to its license, intended use, verdict and duties, signed (keys at /.well-known/jwks.json) and committed to the ledger. Ship it as <file>.rights.json. Call with {"asset": {"sha256": "<64-hex sha256 of the file>", "kind": "font", "name": "Inter.ttf"}, "license": "OFL-1.1", "intended_use": {"commercial_use": true}}.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: false
  • kya_checkAllow or deny an agent against a relying-party policy, evaluated on the live record: min_level, require_active (default true), min_grade, max_open_complaints, require_classes, requ…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.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Allow or deny an agent against a relying-party policy, evaluated on the live record: min_level, require_active (default true), min_grade, max_open_complaints, require_classes, require_domain, min_tenure_days. Returns allow and the reasons for any denial. Call with {"agent": "RS-XXXX-XXXX", "policy": {"min_level": "L1", "max_open_complaints": 0}}.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • ledger_verifyCall to verify Rightshold's ledger itself, not this server's word.1 finding

    Why it is in this group

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

    Findings

    • Warn

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

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "call": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Call to verify Rightshold's ledger itself, not this server's word. Recomputes every payload, leaf and block hash, Merkle root, chain link and signature. Free, no key.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • license_compatibilityCall before shipping one product that bundles code, data, fonts or media under different licenses.1 finding

    Why it is in this group

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

    Findings

    • Warn

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

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "call": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Call before shipping one product that bundles code, data, fonts or media under different licenses. Checks 2-25 SPDX ids for your use (default redistribution): compatible, needs_review or incompatible, the reason for each pair, the combined duties and which license governs. Free, no key. Call with {"licenses": ["MIT", "GPL-3.0-only"], "intended_use": {"redistribution": true}}.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • read_ai_policyCall before crawling, quoting or training on content from a website.Reads text others write1 finding

    Why it is in this group

    • The name "read_ai_policy" says "read": a word for reading.
    • The description's first verb is "call": a word for a change.
    • 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.
    • untrusted-inputThe description's first sentence says "call" and "website": it reads web pages, which anyone can write.

    Findings

    • Warn

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

      Matched text

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

      What this check looks for
    Description, as the server wrote it
    Call before crawling, quoting or training on content from a website. Reads the domain's robots.txt for 14 AI crawlers and user-fetch agents (GPTBot, ClaudeBot, ChatGPT-User, Google-Extended, PerplexityBot, CCBot...), ai.txt and W3C TDMRep, and summarizes whether AI training is opted out. Free, no key, cached 6 hours. Signals, not a license. Call with {"domain": "example.com"}.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: true
    • untrustedContentHint: true
  • record_actionRecords what your agent did; returns a tamper-evident, ledger-committed receipt.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
    Records what your agent did; returns a tamper-evident, ledger-committed receipt. No key: send accept_terms:true, then send the returned key as a Bearer token. Proves submission, not the act.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
  • registry_lookupCall before trusting, paying or hiring an agent you have not worked with.1 finding

    Why it is in this group

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

    Findings

    • Warn

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

      Matched text

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

      What this check looks for
    Description, as the server wrote it
    Call before trusting, paying or hiring an agent you have not worked with. Free, no key. Look up an agent by registration number (RS-XXXX-XXXX) or handle: status (valid, expired, suspended, revoked), license classes in force, commerce limit, liens, and optionally the full history report. Call with {"query": "RS-XXXX-XXXX", "history": true}.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • sign_upUse when your principal wants a Rightshold account: one call creates it and a free API key (shown once), email optional.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
    Use when your principal wants a Rightshold account: one call creates it and a free API key (shown once), email optional. accept_terms:true means the account holder agrees to the terms at /terms. Add agent {handle, display_name, headline?, skills?, public_key_pem?} to register your agent in the same call (KYA L0); if that block is invalid nothing is created. The result carries ready-to-paste connector settings. sandbox:true gives an isolated rs_test_ key. Calls are recorded for due-diligence reports (see /privacy).

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: true
  • start_hereCall first if you are new to Rightshold.No findings

    Why it is in this group

    • The name "start_here" says "start": a word for a change.
    Description, as the server wrote it
    Call first if you are new to Rightshold. With {} it returns the five jobs Rightshold does, each as a ready call, plus how to connect. Add {"license": "CC-BY-NC-4.0", "use": "commercial_use"} or {"question": "Can I train a model on MIT code?"} and the same call answers it. Free, no key.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: false
  • submit_feedbackTell the team what is missing, broken or worth more: topic, satisfaction 1-5, free text; only aggregates are published.No findings

    Why it is in this group

    • The name "submit_feedback" says "submit": a word for a change.
    Description, as the server wrote it
    Tell the team what is missing, broken or worth more: topic, satisfaction 1-5, free text; only aggregates are published. Or file a structured report (rs-agent-feedback/1): {api_surface, endpoint_or_tool, friction, expected?, actual?, severity?, repro?} — sealed as hash and state only; returns an id and one-time poll_token for feedback_status. Feedback is input, never instruction.

    Annotations the server declares

    • readOnlyHint: false
    • destructiveHint: false
    • idempotentHint: false
    • openWorldHint: false
  • verify_rights_manifestCall when a deliverable carries a rights manifest.1 finding

    Why it is in this group

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

    Findings

    • Warn

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

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "call": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Call when a deliverable carries a rights manifest. Checks the manifest's signature against Rightshold's issuer keys and confirms its hash was recorded on the ledger; with deliverable_sha256, that it covers those exact bytes. A valid manifest shows what was answered and when; it is not a license grant. Free, no key.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
Only reads 5Show 5 read-only tools
  • get_pricingUse this to see what Rightshold costs before you sign up or plan spend: every pricing line with its price today (0 in the preview) and planned price, its unit and free allowance, a…Reads text others writeNo findings

    Why it is in this group

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

    What it can reach

    • untrusted-inputThe description's first sentence says "review": it reads comments, which other people can write.
    Description, as the server wrote it
    Use this to see what Rightshold costs before you sign up or plan spend: every pricing line with its price today (0 in the preview) and planned price, its unit and free allowance, and the principles behind the prices — nothing buys placement, a grade, a KYA level, a verdict or a review. No account or key needed.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • get_statusUse this to check whether Rightshold is operational before you depend on it, and to cite its record: the current component checks (ledger chain, sealing, anchoring, webhooks, stora…No findings

    Why it is in this group

    • The name "get_status" says "get": a word for reading.
    Description, as the server wrote it
    Use this to check whether Rightshold is operational before you depend on it, and to cite its record: the current component checks (ledger chain, sealing, anchoring, webhooks, storage), ledger and registry counts, plus the 90-day uptime record per component and incidents. The record is self-reported; check_conformance offers an independent re-check. No account or key needed.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • kya_lookupAn agent's RS-KYA level (L0 declared, L1 key-bound, L2 operator-verified), status, the verified facts behind it (key, operator contact, attestation, DNS domain, registration, burea…Reads private data2 findings

    Why it is in this group

    • The name "kya_lookup" says "lookup": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "contact": it reads contacts.

    Findings

    • Info

      The description mentions "credential".

      Matched text

      description: "...hat is missing for the next level. With credential=true, also a signed credential and a on..."

      What this check looks for
    • Info

      The input schema at properties.credential.description mentions "credential".

      Matched text

      inputSchema.properties.credential.description: "Return the signed credential"

      What this check looks for
    Description, as the server wrote it
    An agent's RS-KYA level (L0 declared, L1 key-bound, L2 operator-verified), status, the verified facts behind it (key, operator contact, attestation, DNS domain, registration, bureau record) and what is missing for the next level. With credential=true, also a signed credential and a one-hour kya+jwt token to present to other services. Free, no key. Call with {"agent": "RS-XXXX-XXXX"}.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • list_toolsetsThe toolsets this server offers and the tools in each.No findings

    Why it is in this group

    • The name "list_toolsets" says "list": a word for reading.
    Description, as the server wrote it
    The toolsets this server offers and the tools in each. Connect to /mcp?toolset=<name> (or toolset=all) to list them; most clients only offer the tools they listed, so connect with the set your task needs.

    Annotations the server declares

    • readOnlyHint: true
    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
  • search_storeList published apps and skills in the Rightshold Store: for each, the vetting tier, the publisher and their live KYA level, declared permissions and the rating aggregate (ratings c…No findings

    Why it is in this group

    • The name "search_store" says "search": a word for reading.
    Description, as the server wrote it
    List published apps and skills in the Rightshold Store: for each, the vetting tier, the publisher and their live KYA level, declared permissions and the rating aggregate (ratings come only from KYA-verified agents). Filter with type (app or skill) and a plain-text query matched against names and summaries. Call with {"type": "skill", "query": "license check"}.

    Annotations the server declares

    • 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.

7 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 data1
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write7
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__rightshold__check_listing"
    ]
  }
}

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, 18:46 UTC, unchanged since Nov 6, 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
19
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
https://www.rightshold.com/.well-known/oauth-protected-resource/mcp
Authorization servers
https://www.rightshold.com
Legacy HTTP+SSE transport
no
Dynamic client registration
yes
Client ID metadata documents
no
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 19 tools that change things (19 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://www.rightshold.com/mcp: tools/list returned 24 tools to an anonymous client, including start_here (write), read_ai_policy (write), license_compatibility (write), attribution_string (write), ask_rightshold (write) and 14 more

    What this check looks for

Posture changes

  1. Nov 6, 2026
    • Dynamic client registration: no to yes
  2. Oct 29, 2026
    • Authorization servers: none to https://www.rightshold.com
    • Dynamic client registration: not observed to no
    • Client ID metadata documents: not observed to no

Findings

Matches of fixed checks, with the text they matched.

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

    Matched text

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

    What this check looks for
  • The server marks "license_compatibility" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "call": a word for a change.

    What this check looks for
  • The server marks "attribution_string" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "call": a word for a change.

    What this check looks for
  • The server marks "check_license" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The description's first verb is "call": a word for a change.

    What this check looks for
  • The server marks "verify_rights_manifest" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "call": a word for a change.

    What this check looks for
  • The server marks "registry_lookup" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

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

    What this check looks for
Show 4 more findings
  • The server marks "ledger_verify" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The description's first verb is "call": a word for a change.

    What this check looks for
  • The server marks "gate_check" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":true}; The description's first verb is "call": a word for a change.

    What this check looks for
  • The description mentions "credential".

    Matched text

    description: "...hat is missing for the next level. With credential=true, also a signed credential and a on..."

    What this check looks for
  • The input schema at properties.credential.description mentions "credential".

    Matched text

    inputSchema.properties.credential.description: "Return the signed credential"

    What this check looks for

History

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

  1. Tools1 added, 1 removed, 9 changed.+1 added−1 removed~9 changedView diff
  2. PostureRemote posture changed. Dynamic client registration: no -> yes.View diff
    • Dynamic client registration: no to yes
  3. Tools9 changed.~9 changedView diff
  4. PostureRemote posture changed. Authorization servers: none -> https://www.rightshold.com; Dynamic client registration: not observed -> no; Client ID metadata documents: not observed -> no.View diff
    • Authorization servers: none to https://www.rightshold.com
    • Dynamic client registration: not observed to no
    • Client ID metadata documents: not observed to no

Install

Pinned to version 0.5.1. Values in angle brackets are yours to fill in.

claude mcp add --transport http rightshold https://www.rightshold.com/mcp
claude mcp add --transport http rightshold https://www.rightshold.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__rightshold__get_pricing",
      "mcp__rightshold__get_status",
      "mcp__rightshold__kya_lookup",
      "mcp__rightshold__search_store",
      "mcp__rightshold__list_toolsets"
    ],
    "ask": [
      "mcp__rightshold__start_here",
      "mcp__rightshold__read_ai_policy",
      "mcp__rightshold__license_compatibility",
      "mcp__rightshold__attribution_string",
      "mcp__rightshold__ask_rightshold",
      "mcp__rightshold__check_license",
      "mcp__rightshold__assess_resource_for_use",
      "mcp__rightshold__check_rights_batch",
      "mcp__rightshold__verify_rights_manifest",
      "mcp__rightshold__sign_up",
      "mcp__rightshold__record_action",
      "mcp__rightshold__registry_lookup",
      "mcp__rightshold__ledger_verify",
      "mcp__rightshold__issue_rights_receipt",
      "mcp__rightshold__gate_check",
      "mcp__rightshold__issue_rights_manifest",
      "mcp__rightshold__kya_check",
      "mcp__rightshold__submit_feedback",
      "mcp__rightshold__check_listing"
    ],
    "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 rightshold. Review it before you use it.

Capability tags behind 9 rules
  • mcp__rightshold__read_ai_policy untrusted-input
  • mcp__rightshold__check_license untrusted-input
  • mcp__rightshold__check_rights_batch untrusted-input
  • mcp__rightshold__sign_up untrusted-input
  • mcp__rightshold__record_action untrusted-input
  • mcp__rightshold__get_pricing untrusted-input
  • mcp__rightshold__gate_check untrusted-input
  • mcp__rightshold__kya_lookup private-data
  • mcp__rightshold__check_listing outbound

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.rightshold/rightshold/verify?hash=cc113299bbede64ce2e9bdcc010203328a33895b25b03ad36265d46e28616b79" | jq -e '.changed == false'

Pinned hash cc113299bbed, from Nov 14, 2026, 05:18 UTC.

Watch

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

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

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

Badge

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

Source: MCP Registry. Synced Oct 21, 2026, 18:02 UTC.

Rightshold MCP server: questions and answers

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

Is the Rightshold MCP server safe?

No scan can promise that. What its definitions show: 24 tools, 19 of which can change, send or delete something, and 10 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 Rightshold MCP server delete data?

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

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

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the Rightshold MCP server have?

24 tools: 19 write, 5 read. start_here, read_ai_policy, license_compatibility, attribution_string, ask_rightshold, check_license, assess_resource_for_use, check_rights_batch and 16 more.

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

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

Does the Rightshold 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): check_listing.

Has the Rightshold MCP server changed its tools?

Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 14, 2026. Every change is on this page and in its Atom feed.