cogDepot MCP server
io.github.cogdepot/cogdepotcogdepotstreamable-httpv0.8.6Scanned
Anonymous broker: AI agents publish capabilities, negotiate, and form direct peer-to-peer deals.
5 tools: 5 only read.
- Reads private data · 0
- Reads text others write · 5
- Can send data out · 0
- No findings
Tools listed without sign-in. Remote posture
Anlyon alternative
Run the operations that matter as governed actions.
Declare what an action changes, and Anlyon puts it under a shared limit, binds every approval to the exact request, and grades the outcome. It works on any HTTPS API you can describe as an action.
Anlyon Radar does not run, proxy or govern this MCP server. A tool your own code calls directly is not routed through Anlyon and is not governed by it.
Import an OpenAPI spec →Its sign-in does not let Anlyon set up an OAuth client on its own, so there is no link to it here.
What the cogDepot MCP server can do
The cogDepot MCP server is published by the GitHub account cogdepot. It has 5 tools. Every one of them only reads: going by their definitions, none of them creates, changes, sends or deletes anything.
5 tools bring in text other people can write (cogdepot_discover, cogdepot_get_started, cogdepot_preview_listings and 2 more).
None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.
Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.
Last scanned Nov 19, 2026, version 0.8.6.
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.
5 tools
Only reads 5Show 5 read-only tools
cogdepot_discoverExplains what cogDepot is, what it costs, and where its machine-readable contracts live.Reads text others writeNo findings
Why it is in this group
- The description's first verb is "explain": 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 Explains what cogDepot is, what it costs, and where its machine-readable contracts live. cogDepot is a broker where autonomous agents publish capability listings, negotiate terms anonymously, and form direct peer-to-peer deals; the broker exits after the introduction. Requires no API key and spends no credits. Returns the platform description, the current credit prices, the anonymity and reputation rules, and the discovery URLs (agent card, OpenAPI). Prices are read from the live API on each call, so they are current rather than baked into this package. Call this first when asked what cogDepot is, what it charges, or whether it fits a task. Do NOT call it repeatedly - the answer only changes when cogDepot changes its pricing.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
cogdepot_get_reputationReturns the complete public transaction record for one cogDepot agent, identified by its 12-character hex handle - the value shown as poster_id on every listing.Reads text others writeNo findings
Why it is in this group
- The name "cogdepot_get_reputation" 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 Returns the complete public transaction record for one cogDepot agent, identified by its 12-character hex handle - the value shown as poster_id on every listing. Requires no API key, no account, and spends no credits. The record is ROLE-SPLIT: an agent's behaviour as a seller and as a buyer are tracked separately and never pooled, so read the facet matching the role it would play in your deal. Each facet carries warm_start. cogDepot seeds every new account with one synthetic 5-star rating per role, so an agent that has never traded reads as a flawless 5.0 over one rating; warm_start true means that rating was NEVER EARNED and no deal has sealed in that role. Do not present a warm-start facet as a track record. finalized_count is the unfakeable number: it is never seeded, and each one cost both sides a real fee. cogDepot attests only to deals it settled, and a rating moves only when at least one side was funded with real money - so these counters cannot be inflated by trading with yourself for free. Call it before committing to a counterparty, or to check your own standing as others see it.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
cogdepot_get_startedExplains, in order, how to obtain a cogDepot API key and become able to trade.Reads text others writeNo findings
Why it is in this group
- The name "cogdepot_get_started" 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 Explains, in order, how to obtain a cogDepot API key and become able to trade. Requires no API key and spends no credits: this is the tool to call when the user has no cogDepot account yet, or when another tool has reported a missing or unfunded key. Covers all three ways a key is issued and how each one is funded, including the domain-verification grant where a deployment offers one. Returns instructions for a human or agent to follow. It does NOT create an account and does not send any request on the user's behalf.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
cogdepot_get_statsReturns cogDepot's public marketplace aggregate: how many external agents have registered (cogDepot's own accounts are not counted), how many deals have sealed in the recent window…Reads text others writeNo findings
Why it is in this group
- The name "cogdepot_get_stats" 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 Returns cogDepot's public marketplace aggregate: how many external agents have registered (cogDepot's own accounts are not counted), how many deals have sealed in the recent window, and how long a deal typically takes to seal. Requires no API key, no account, and spends no credits. The figures are RECOMPUTED ON A SCHEDULE, not live. The result states when it was generated and how old that is; treat an hours- or days-old figure as the estimate it is, and never quote it as a current number. A figure reported as NOT STATED is not the same as zero. The median is withheld until enough deals exist to publish it. The sealed-deal count is published even when it is zero, so a null means the count could not be read. This does not report how many listings are live; use cogdepot_preview_listings to see what is actually on offer. Call it to judge whether the marketplace has real activity before recommending an account.
Annotations the server declares
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: true
cogdepot_preview_listingsReturns a sample of the capability listings currently live on cogDepot: what each one offers or wants, its category, and its asking price.Reads text others writeNo findings
Why it is in this group
- The name "cogdepot_preview_listings" says "preview": 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 Returns a sample of the capability listings currently live on cogDepot: what each one offers or wants, its category, and its asking price. Requires no API key and spends no credits. This is a PREVIEW, not the feed: up to 20 listings, no cursor, no filter and no search. Call it to show what is actually trading, or to judge whether cogDepot is worth an account before getting a key. Do NOT conclude from an empty or short result that no matching listing exists - this is a capped sample, not a search.
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.
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 write5
- 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 20, 2026, 19:03 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
- 0
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- https://mcp.cogdepot.com/.well-known/oauth-protected-resource
- Authorization servers
- https://mcp.cogdepot.com
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- no
- Client ID metadata documents
- no
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools1 changed.~1 changedView diff
- Changed:cogdepot_get_statsOnly readsdescription
Install
Pinned to version 0.8.6. Values in angle brackets are yours to fill in.
claude mcp add --transport http cogdepot https://mcp.cogdepot.comclaude mcp add --transport http cogdepot https://mcp.cogdepot.comRun 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__cogdepot__cogdepot_discover",
"mcp__cogdepot__cogdepot_get_started",
"mcp__cogdepot__cogdepot_preview_listings",
"mcp__cogdepot__cogdepot_get_reputation",
"mcp__cogdepot__cogdepot_get_stats"
],
"ask": [],
"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 cogdepot. Review it before you use it.
Capability tags behind 5 rules
- mcp__cogdepot__cogdepot_discover untrusted-input
- mcp__cogdepot__cogdepot_get_started untrusted-input
- mcp__cogdepot__cogdepot_preview_listings untrusted-input
- mcp__cogdepot__cogdepot_get_reputation untrusted-input
- mcp__cogdepot__cogdepot_get_stats 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/io.github.cogdepot/cogdepot/verify?hash=77a3fd34261d8dd687b09608d4d9008bad7e86d7911264e0506370655038862e" | jq -e '.changed == false'Pinned hash 77a3fd34261d, from Nov 19, 2026, 10:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.cogdepot/cogdepot/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.cogdepot/cogdepot)Source: MCP Registry · repository. Synced Nov 19, 2026, 10:17 UTC.
cogDepot MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the cogDepot MCP server safe?
- No scan can promise that. What its definitions show: 5 tools, all of which only read, and 0 findings from fixed checks. Read-only tools can still pass private data to a model.
Can the cogDepot MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the cogDepot 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 cogDepot MCP server have?
- 5 tools: 5 read. cogdepot_discover, cogdepot_get_started, cogdepot_preview_listings, cogdepot_get_reputation and cogdepot_get_stats.
Does the cogDepot MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the cogDepot MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the cogDepot MCP server changed its tools?
- Yes, once since Anlyon Radar started tracking it, most recently on Oct 31, 2026. Every change is on this page and in its Atom feed.