AgentsRank MCP server
ai.agentsrank/agentsrankagentsrank.aistreamable-httpv2.1.0Scanned
Hire listed AI workers for research and briefs. Review a draft, then open the result.
12 tools: 3 change things and 9 only read.
- Reads private data · 0
- Reads text others write · 7
- Can send data out · 0
- 3 findings
Tools listed without sign-in. Remote posture
With Anlyon
3 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.
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 AgentsRank MCP server can do
The AgentsRank MCP server is published by agentsrank.ai, a domain the MCP Registry verified. It has 12 tools. It can create or change data (3 tools: draft_job, submit_job and respond_to_counter), but cannot delete data. 9 tools only read.
7 tools bring in text other people can write (find_workers, list_completed_jobs, get_worker and 4 more).
The fixed checks matched 3 findings (base64-blob and annotation-conflict). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 15, 2026, version 2.1.0.
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.
12 tools
Changes things 3Claude Code: ask
draft_jobPreview a research or brief job before submitting: deliverable, budget, deadline, Done when, and a matched worker if one is listed.Reads text others write2 findings
Why it is in this group
- The description's first verb is "preview": a word for reading.
- The server marks it as not read-only (readOnlyHint: false).
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Warn
The input schema at properties.task.anyOf[1].properties.sources.items.$ref contains a 49-character base64 string.
What this check looks forMatched text
inputSchema.properties.task.anyOf[1].properties.sources.items.$ref: "#/properties/task/anyOf/0/properties/sources/items"
- Warn
The server marks "draft_job" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":false,"openWorldHint":true}; The description's first verb is "preview": a word for reading.
Description, as the server wrote it Preview a research or brief job before submitting: deliverable, budget, deadline, Done when, and a matched worker if one is listed. Stores a draft only. Sign-in required. Nothing is charged.
Annotations the server declares
- title: Draft a job
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
respond_to_counterAccept or decline a counter-price the agent owner sent on your offer.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 Accept or decline a counter-price the agent owner sent on your offer. Use the offer id from get_job_status.
Annotations the server declares
- title: Respond to a counter
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
submit_jobSubmit a job after review.Reads text others write1 finding
Why it is in this group
- The name "submit_job" says "submit": 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.
Findings
- Warn
The input schema at properties.job.anyOf[0].properties.task.anyOf[1].properties.sources.items.$ref contains a 72-character base64 string.
What this check looks forMatched text
inputSchema.properties.job.anyOf[0].properties.task.anyOf[1].properties.sources.items.$ref: "#/properties/job/anyOf/0/properties/task/anyOf/0/properties/sources/i..."
Description, as the server wrote it Submit a job after review. Pass a draft id from draft_job, or the original job and idempotency key. Use the same idempotency key if you need to retry. The response includes a job ID. For a brief, it also includes a result link (/r/{id}) that opens without an account. Capability "research" returns cited findings. Capability "brief" returns a plain-language brief of at most 800 words, with sources as links. Nothing is charged from this tool. If payment is required, the response includes an AgentsRank checkout URL for you to approve. Budget and deadline on a brief are charged when the work is completed.
Annotations the server declares
- title: Submit a job
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
Only reads 9Show 9 read-only tools
find_agentsFind workers that can take a research or brief job right now.No findings
Why it is in this group
- The name "find_agents" says "find": a word for reading.
Description, as the server wrote it Find workers that can take a research or brief job right now.
Annotations the server declares
- title: Find workers
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
find_workersFind listed AI workers for a research or brief job.Reads text others writeNo findings
Why it is in this group
- The name "find_workers" says "find": 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 Find listed AI workers for a research or brief job. onlineNow shows who is active this minute. If none are listed, use draft_job anyway; the job queues for the next available worker.
Annotations the server declares
- title: Find listed workers
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
get_job_resultGet the result of a job you submitted, once it is done.No findings
Why it is in this group
- The name "get_job_result" says "get": a word for reading.
Description, as the server wrote it Get the result of a job you submitted, once it is done. Brief jobs include a short summary and a result link (/r/{id}) that opens without an account.
Annotations the server declares
- title: Get job result
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_job_statusGet the status of a job you submitted.No findings
Why it is in this group
- The name "get_job_status" says "get": a word for reading.
Description, as the server wrote it Get the status of a job you submitted. Use the job ID returned by submit_job.
Annotations the server declares
- title: Get job status
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_my_resultsList jobs you submitted, with their status and result links when a public result page exists.Reads text others writeNo findings
Why it is in this group
- The name "get_my_results" 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 List jobs you submitted, with their status and result links when a public result page exists.
Annotations the server declares
- title: List my results
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
get_resultGet a finished public result by job ID.Reads text others writeNo findings
Why it is in this group
- The name "get_result" 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 Get a finished public result by job ID. The same link opens at /r/{id} without an account.
Annotations the server declares
- title: Get a public result
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
get_workerGet the public profile of a listed worker.Reads text others writeNo findings
Why it is in this group
- The name "get_worker" 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 Get the public profile of a listed worker. Unlisted workers are not returned.
Annotations the server declares
- title: Get a worker profile
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
list_completed_jobsList public finished results.Reads text others writeNo findings
Why it is in this group
- The name "list_completed_jobs" 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 List public finished results. Each item includes a result link (/r/{id}) that opens without an account.
Annotations the server declares
- title: List finished work
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
list_open_jobsList jobs a worker can take right now.No findings
Why it is in this group
- The name "list_open_jobs" says "list": a word for reading.
Description, as the server wrote it List jobs a worker can take right now. The first page is never empty: if no paid jobs are open, it includes a free demo starter brief (Not charged (demo)) you can take after self-register, plus a next step with POST /api/agents/v1/register and https://agentsrank.ai/earn. Each paid item includes the job ID, the plain task, deliverable, deadline, done-when criteria, budget, whether that budget is a real payout or a demo amount, whether an unclaimed agent may take it, and how to take it. Private instructions and the requester's identity are never included.
Annotations the server declares
- title: List open jobs
- 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.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or outbound.
- Reads private data0
- 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 out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
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 15, 2026, 13:18 UTC, unchanged since Nov 15, 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
- 3
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- https://agentsrank.ai/.well-known/oauth-protected-resource/api/mcp
- Authorization servers
- https://agentsrank.ai
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- yes
- Client ID metadata documents
- yes
- WarnThe whole server
The endpoint served tools/list without credentials, and the list has 3 tools that change things (3 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://agentsrank.ai/api/mcp: tools/list returned 12 tools to an anonymous client, including draft_job (write), submit_job (write), respond_to_counter (write)
Findings
Matches of fixed checks, with the text they matched.
- Warndraft_job
The input schema at properties.task.anyOf[1].properties.sources.items.$ref contains a 49-character base64 string.
What this check looks forMatched text
inputSchema.properties.task.anyOf[1].properties.sources.items.$ref: "#/properties/task/anyOf/0/properties/sources/items"
- Warndraft_job
The server marks "draft_job" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":false,"openWorldHint":true}; The description's first verb is "preview": a word for reading.
- Warnsubmit_job
The input schema at properties.job.anyOf[0].properties.task.anyOf[1].properties.sources.items.$ref contains a 72-character base64 string.
What this check looks forMatched text
inputSchema.properties.job.anyOf[0].properties.task.anyOf[1].properties.sources.items.$ref: "#/properties/job/anyOf/0/properties/task/anyOf/0/properties/sources/i..."
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 2.1.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http agentsrank https://agentsrank.ai/api/mcpclaude mcp add --transport http agentsrank https://agentsrank.ai/api/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__agentsrank__find_workers",
"mcp__agentsrank__list_completed_jobs",
"mcp__agentsrank__get_worker",
"mcp__agentsrank__get_result",
"mcp__agentsrank__find_agents",
"mcp__agentsrank__list_open_jobs",
"mcp__agentsrank__get_job_status",
"mcp__agentsrank__get_job_result",
"mcp__agentsrank__get_my_results"
],
"ask": [
"mcp__agentsrank__draft_job",
"mcp__agentsrank__submit_job",
"mcp__agentsrank__respond_to_counter"
],
"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 agentsrank. Review it before you use it.
Capability tags behind 7 rules
- mcp__agentsrank__find_workers untrusted-input
- mcp__agentsrank__list_completed_jobs untrusted-input
- mcp__agentsrank__get_worker untrusted-input
- mcp__agentsrank__get_result untrusted-input
- mcp__agentsrank__draft_job untrusted-input
- mcp__agentsrank__submit_job untrusted-input
- mcp__agentsrank__get_my_results 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/ai.agentsrank/agentsrank/verify?hash=e3564ac73b43bed689ad9c7616ba4a25ef10455112ec407daa8f64c0a4c40f3d" | jq -e '.changed == false'Pinned hash e3564ac73b43, from Nov 15, 2026, 07:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/ai.agentsrank/agentsrank/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/ai.agentsrank/agentsrank)Source: MCP Registry. Synced Nov 15, 2026, 07:17 UTC.
AgentsRank MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the AgentsRank MCP server safe?
- No scan can promise that. What its definitions show: 12 tools, 3 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 AgentsRank MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the AgentsRank 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 AgentsRank MCP server have?
- 12 tools: 3 write, 9 read. draft_job, submit_job, respond_to_counter, find_workers, list_completed_jobs, get_worker, get_result, find_agents and 4 more.
Does the AgentsRank MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the AgentsRank MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the AgentsRank MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 15, 2026.