Agent Service Directory MCP server
com.agentservicedirectory/agent-service-directoryagentservicedirectory.comstreamable-httpv0.2.0Scanned
A directory of agent services, with metrics.
10 tools: 1 do something that can't be undone, 3 change things and 6 only read.
- Reads private data · 1
- Reads text others write · 0
- Can send data out · 2
- 5 findings
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.
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 Agent Service Directory MCP server can do
The Agent Service Directory MCP server is published by agentservicedirectory.com, a domain the MCP Registry verified. It has 10 tools. It can do things that cannot be undone, such as sending, paying or publishing (send_feedback) and create or change data (3 tools: get_service, report_outcome and submit_service), but cannot delete data. 6 tools only read.
1 tool reads private data (get_tree); 2 tools can send data out of your system (send_feedback and submit_service).
The fixed checks matched 5 findings (credentials-mention and annotation-conflict). 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 12, 2026, version 0.2.0.
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.
10 tools
Can't be undone 1Claude Code: ask
send_feedbackTell us what is missing or wrong: a service or kind of service we don't list, a metric you need, a feature, wrong data or unclear docs.Can send data outNo findings
Why it is in this group
- The name "send_feedback" says "send": a word for sending, and what is sent cannot be taken back.
What it can reach
- outboundThe name "send_feedback" says "send": it sends or publishes something to other people.
Description, as the server wrote it Tell us what is missing or wrong: a service or kind of service we don't list, a metric you need, a feature, wrong data or unclear docs. No key needed. We use it to decide where to look next; counts by kind are public on the roadmap, the messages are not. Never include task data, secrets or personal information.
Annotations the server declares
- readOnlyHint: false
- idempotentHint: false
- openWorldHint: false
Changes things 3Claude Code: ask
get_serviceFull evidence and connection details for one service, plus a referral_id.1 finding
Why it is in this group
- The name "get_service" says "get": a word for reading.
- The server marks it as not read-only (readOnlyHint: false).
Findings
- Warn
The server marks "get_service" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"idempotentHint":false,"openWorldHint":false}; The name "get_service" says "get": a word for reading.
Description, as the server wrote it Full evidence and connection details for one service, plus a referral_id. Use the referral_id with report_outcome after you use the service (valid 7 days).
Annotations the server declares
- readOnlyHint: false
- idempotentHint: false
- openWorldHint: false
report_outcomeAfter using a service, report whether it worked.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 After using a service, report whether it worked. Requires a free API key (Authorization: Bearer asd_...; get one with POST https://agentservicedirectory.com/api/v1/keys). Quote the referral_id from get_service. Reports are stored as unverified claims and published only after moderation. Never include task data, secrets or personal information.
Annotations the server declares
- readOnlyHint: false
- idempotentHint: true
- openWorldHint: false
submit_serviceGive us the address of a paid API endpoint (one that answers with HTTP 402), a remote MCP server or an A2A agent card that we don't list yet.Can send data outNo findings
Why it is in this group
- The name "submit_service" says "submit": a word for a change.
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 Give us the address of a paid API endpoint (one that answers with HTTP 402), a remote MCP server or an A2A agent card that we don't list yet. No key needed. We check it like a service from a registry: one unpaid request per scheduled interval, never a paid call, no tool called. It joins the directory if its checks meet our listing criteria. We keep only the address; returns the service id and page.
Annotations the server declares
- readOnlyHint: false
- idempotentHint: true
- openWorldHint: false
Only reads 6Show 6 read-only tools
get_changesRecently detected changes to monitored documentation and API schemas, across all reviewed services or for one.No findings
Why it is in this group
- The name "get_changes" says "get": a word for reading.
Description, as the server wrote it Recently detected changes to monitored documentation and API schemas, across all reviewed services or for one. A detected change is pending interpretation, not necessarily a breaking change.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_historyIndividual check results for a service over the last days, newest first.No findings
Why it is in this group
- The name "get_history" says "get": a word for reading.
Description, as the server wrote it Individual check results for a service over the last days, newest first. Use it to see patterns (intermittent failures, slowdowns) behind the summary figures.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_reportThe weekly State of Agent Services report: coverage, probe results, payment networks, prices and purpose tags across everything we track.No findings
Why it is in this group
- The name "get_report" says "get": a word for reading.
Description, as the server wrote it The weekly State of Agent Services report: coverage, probe results, payment networks, prices and purpose tags across everything we track. Counts and distributions only, with their method. Omit id for the latest published report; 'current' gives provisional week-to-date figures.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_statsHow many services we track and how many checks ran in the last 24 hours.No findings
Why it is in this group
- The name "get_stats" says "get": a word for reading.
Description, as the server wrote it How many services we track and how many checks ran in the last 24 hours.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
get_treeThe capability tree: what services do, as branches, leaves and leaves' children, each with a one-line description and the number of services in it (zeros included).Reads private dataNo findings
Why it is in this group
- The name "get_tree" says "get": a word for reading.
What it can reach
- private-dataThe description's first sentence says "branch": it reads repositories.
Description, as the server wrote it The capability tree: what services do, as branches, leaves and leaves' children, each with a one-line description and the number of services in it (zeros included). Read it once (follow its guide), choose the most specific node that matches your task, then call search_services with branch set to its id, or up to three ids separated by commas when more than one fits. format "text" returns the same tree as plain text, one line per node: the cheapest to read. Keep it and pass since_revision later to get only what changed. IDs are permanent: an id that moved (listed under moved) still works as a branch filter.
Annotations the server declares
- readOnlyHint: true
- openWorldHint: false
search_servicesFind services an agent can use, by what they do: remote MCP servers (with the tools each lists), local MCP servers (packages), pay-per-call APIs (x402, MPP), A2A agents and task ma…4 findings
Why it is in this group
- The name "search_services" says "search": a word for reading.
Findings
- Info
The description mentions "credentials".
What this check looks forMatched text
description: "...r on type, access (none_asked, payment, credentials), max_price_usd and the evidence (passi..."
- Info
The input schema at properties.query.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.query.description: "..., keyless, no sign-in: does not ask for credentials (observed or stated); "an API key is fi..."
- Info
The input schema at properties.access.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.access.description: "...ent' (our plain request got HTTP 402), 'credentials' (HTTP 401, or its agent card or MCP Re..."
- Info
The input schema at properties.access.enum[1] mentions "credentials".
What this check looks forMatched text
inputSchema.properties.access.enum[1]: "credentials"
Description, as the server wrote it Find services an agent can use, by what they do: remote MCP servers (with the tools each lists), local MCP servers (packages), pay-per-call APIs (x402, MPP), A2A agents and task marketplaces, with our checks for each. Search by words (query) or by capability tree node (branch); filter on type, access (none_asked, payment, credentials), max_price_usd and the evidence (passing, min_pass_rate); then call get_service for connection details. Results are compact; use limit/offset to page.
Annotations the server declares
- readOnlyHint: 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 untrusted-input.
- Reads private data1
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write0
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out2
- 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, 18:33 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
- 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.
What this check looks forMatched text
https://agentservicedirectory.com/mcp: tools/list returned 10 tools to an anonymous client, including send_feedback (irreversible), get_service (write), report_outcome (write), submit_service (write)
Posture changes
- Nov 6, 2026
- Tools that change things, listed without credentials: 3 to 4
Findings
Matches of fixed checks, with the text they matched.
- Warnget_service
The server marks "get_service" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"idempotentHint":false,"openWorldHint":false}; The name "get_service" says "get": a word for reading.
- Infosearch_services
The description mentions "credentials".
What this check looks forMatched text
description: "...r on type, access (none_asked, payment, credentials), max_price_usd and the evidence (passi..."
- Infosearch_services
The input schema at properties.query.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.query.description: "..., keyless, no sign-in: does not ask for credentials (observed or stated); "an API key is fi..."
- Infosearch_services
The input schema at properties.access.description mentions "credentials".
What this check looks forMatched text
inputSchema.properties.access.description: "...ent' (our plain request got HTTP 402), 'credentials' (HTTP 401, or its agent card or MCP Re..."
- Infosearch_services
The input schema at properties.access.enum[1] mentions "credentials".
What this check looks forMatched text
inputSchema.properties.access.enum[1]: "credentials"
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools1 added, 1 changed.+1 added~1 changedView diff
- Added:submit_serviceChanges things
- Changed:search_servicesOnly readsinputSchema
PostureRemote posture changed. Tools that change things, listed without credentials: 3 -> 4.View diff
- Tools that change things, listed without credentials: 3 to 4
Tools1 changed.~1 changedView diff
- Changed:search_servicesOnly readsinputSchema
Install
Pinned to version 0.2.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http agent-service-directory https://agentservicedirectory.com/mcpclaude mcp add --transport http agent-service-directory https://agentservicedirectory.com/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__agent-service-directory__search_services",
"mcp__agent-service-directory__get_history",
"mcp__agent-service-directory__get_changes",
"mcp__agent-service-directory__get_tree",
"mcp__agent-service-directory__get_stats",
"mcp__agent-service-directory__get_report"
],
"ask": [
"mcp__agent-service-directory__get_service",
"mcp__agent-service-directory__report_outcome",
"mcp__agent-service-directory__send_feedback",
"mcp__agent-service-directory__submit_service"
],
"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 agent-service-directory. Review it before you use it.
Capability tags behind 3 rules
- mcp__agent-service-directory__get_tree private-data
- mcp__agent-service-directory__send_feedback outbound
- mcp__agent-service-directory__submit_service 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.agentservicedirectory/agent-service-directory/verify?hash=a22640f34cd72fc5a4cf33b343976fc7302b7d60e5dc971ab7a5a5a401567f34" | jq -e '.changed == false'Pinned hash a22640f34cd7, from Nov 12, 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.agentservicedirectory/agent-service-directory/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/com.agentservicedirectory/agent-service-directory)Source: MCP Registry. Synced Oct 21, 2026, 17:48 UTC.
Agent Service Directory MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Agent Service Directory MCP server safe?
- No scan can promise that. What its definitions show: 10 tools, 4 of which can change, send or delete something, and 5 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 Agent Service Directory MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Agent Service Directory MCP server do things that cannot be undone?
- Yes. 1 tool can send, pay, publish or merge: send_feedback.
What tools does the Agent Service Directory MCP server have?
- 10 tools: 1 irreversible, 3 write, 6 read. send_feedback, get_service, report_outcome, submit_service, search_services, get_history, get_changes, get_tree and 2 more.
Does the Agent Service Directory MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Agent Service Directory MCP server send data to outside services?
- 2 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): send_feedback and submit_service.
Has the Agent Service Directory MCP server changed its tools?
- Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 12, 2026. Every change is on this page and in its Atom feed.