Skip to content

AgentIndex — measured reliability for agent services MCP server

io.github.cengit1/agent-indexcengit1streamable-httpv1.1.0Scanned

Measured uptime, price and payment evidence for 19,021 agent services. Reliability, not quality.

22 tools: 1 do something that can't be undone, 9 change things and 12 only read.

Tools: 0 delete, 1 irreversible, 9 write, 12 read.
  • Reads private data · 1
  • Reads text others write · 1
  • Can send data out · 0
  • No 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.
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 AgentIndex — measured reliability for agent services MCP server can do

The AgentIndex — measured reliability for agent services MCP server is published by the GitHub account cengit1. It has 22 tools. It can do things that cannot be undone, such as sending, paying or publishing (payment_evidence) and create or change data (9 tools: preflight_check, bulk_score, price_check, cheapest_that_works and 5 more), but cannot delete data. 12 tools only read.

1 tool reads private data (get_rankings); 1 tool brings in text other people can write (get_agent_profile).

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.

Last scanned Nov 16, 2026, version 1.1.0.

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.

22 tools

Can't be undone 1Claude Code: ask
  • payment_evidencePayment evidence.No findings

    Why it is in this group

    • The name "payment_evidence" says "payment": a word that usually means money moves, which cannot be taken back.
    Description, as the server wrote it
    Payment evidence. PRICED: $0.002 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. On-chain payment evidence for one service: how many payments, how many distinct paying wallets, first and last seen. Read off the chain, not off a claim. Counts are per receiving wallet, so one operator on several hostnames shows the same totals on each.
Changes things 9Claude Code: ask
  • bulk_exportCursor-paginated bulk snapshot of endpoints or resources.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
    Cursor-paginated bulk snapshot of endpoints or resources. PRICED: $0.25 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Returns a 1000-record page. Price is per page, not per record. Pass the returned cursor to continue.
  • bulk_scoreBulk scoring.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
    Bulk scoring. PRICED: $0.02 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Post up to 1000 domains, get one scored row back per row in: score, grade, 90-day uptime, probe count and distinct on-chain payers. The shape an enrichment pipeline expects.
  • cheapest_that_worksCheapest that works.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
    Cheapest that works. PRICED: $0.01 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Best value in one capability category: observed price ranked against our own measured uptime. A cheap service that is down is worth nothing, so price alone is not the ranking.
  • gone_darkGone dark.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
    Gone dark. PRICED: $0.005 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Services that used to answer and have stopped. Never-alive services are excluded, so this lists things that broke rather than things that never worked.
  • has_it_changedChange watch.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
    Change watch. PRICED: $0.002 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Has this endpoint changed since a moment you name. Stateless by design: you pass the time you last looked, we compare it against our probe record. No account and no subscription to administer.
  • new_servicesNew services.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
    New services. PRICED: $0.01 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Services that appeared recently, roughly 170 a day. New means new to us; a service may predate our first sight of it.
  • preflight_checkPre-flight check.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
    Pre-flight check. PRICED: $0.001 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Is this endpoint alive, how long we have known it, what protocols it answers on, what it charges, and whether real money has ever reached it. Answered from our own probe record rather than the service's own claims. First 500 calls a day are free.
  • price_checkPrice check.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
    Price check. PRICED: $0.001 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. What this endpoint charges as we last observed it, not as it advertises, with the live payment terms and whether they match the published ones.
  • uptime_historyUptime history.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
    Uptime history. PRICED: $0.002 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. The 90-day probe record for one service, probe by probe. This is the part nobody who started later can reconstruct.
Only reads 12Show 12 read-only tools
  • find_toolTool finder.No findings

    Why it is in this group

    • The name "find_tool" says "find": a word for reading.
    Description, as the server wrote it
    Tool finder. PRICED: $0.002 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Find services by the tools they actually publish, searching the real tool lists of the 6,530 services that expose one rather than their marketing text.
  • get_agent_profileRetrieve a full profile for a specific agent or endpoint by its numeric ID or identifier URI.Reads text others writeNo findings

    Why it is in this group

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

    What it can reach

    • untrusted-inputThe description's first sentence says "retrieve" and "uri": it reads web pages, which anyone can write.
    Description, as the server wrote it
    Retrieve a full profile for a specific agent or endpoint by its numeric ID or identifier URI. Returns protocol detection, liveness history, on-chain payment evidence, MCP tools list, and A2A skills. FREE for the first 50 calls a day per caller, then $0.02 in USDC on base via x402. Over the allowance, call without `payment` to receive the exact payment requirements, then call again with `payment` to be served.
  • get_demand_previewFree preview of the AgentIndex Builder Demand Report: the top three themes builders are asking about right now, with signal counts.No findings

    Why it is in this group

    • The name "get_demand_preview" says "get": a word for reading.
    Description, as the server wrote it
    Free preview of the AgentIndex Builder Demand Report: the top three themes builders are asking about right now, with signal counts. The full report -- evidence, supply gaps, release inventory and recommendations -- is the priced tool get_demand_report.
  • get_demand_reportBuilder Demand Report: what builders are asking for right now, and where demand has no supply.No findings

    Why it is in this group

    • The name "get_demand_report" says "get": a word for reading.
    Description, as the server wrote it
    Builder Demand Report: what builders are asking for right now, and where demand has no supply. PRICED: $0.05 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Themes ranked by how often they recur across a nightly-mined corpus of developer questions, forum discussion and search demand, each with the underlying signals as evidence and a stated implication. Crossed against our own measured index to report supply per theme, so a capability that is heavily asked for and rarely built shows up as a number rather than a hunch. Includes a current-version inventory for the main agent protocol projects. Caveats travel inside the payload.
  • get_endpoint_alertsChange feed of recently added, changed or newly dark endpoints.No findings

    Why it is in this group

    • The name "get_endpoint_alerts" says "get": a word for reading.
    Description, as the server wrote it
    Change feed of recently added, changed or newly dark endpoints. PRICED: $0.05 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Returns up to 500 recent endpoint changes, filterable by protocol and time window. Intended to be called repeatedly in place of re-crawling the index.
  • get_liveness_historyLiveness attestation and retained probe history for one endpoint.No findings

    Why it is in this group

    • The name "get_liveness_history" says "get": a word for reading.
    Description, as the server wrote it
    Liveness attestation and retained probe history for one endpoint. PRICED: $0.01 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Probes the endpoint live at request time and returns that alongside the complete retained probe series -- every observation oldest-first, uptime percentage, observed-day count, and each up/down transition with the timestamp it changed and how long the previous state had held. History is a function of elapsed time and cannot be reconstructed after the fact.
  • get_rankingsThe full ranked table of every scored service.Reads private dataNo findings

    Why it is in this group

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

    What it can reach

    • private-dataThe description's first sentence says "table": it reads databases.
    Description, as the server wrote it
    The full ranked table of every scored service. PRICED: $0.25 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Every service we hold probe history for, scored out of 100, ordered best first, each tagged with the categories it leads on. Page with limit and offset. Services we have never probed are absent rather than ranked last, because no history is an absence of evidence, not a bad verdict.
  • get_service_reportOne-page report on a single indexed service, ending in a verdict.No findings

    Why it is in this group

    • The name "get_service_report" says "get": a word for reading.
    Description, as the server wrote it
    One-page report on a single indexed service, ending in a verdict. PRICED: $0.02 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Live status, 30 and 90 day uptime computed from our own probe history, advertised price, on-chain payment evidence with distinct payer count, TLS issuer and expiry, the jobs the service declares, and three closest alternatives with their own numbers. Closes with safe / risky / dead and the reason. Every figure cites the table it came from. A domain we do not hold is refused free of charge rather than sold an empty report; check first with the free coverage lookup at /covered/{domain}.
  • get_top_tenThe ten best services in one capability category today.No findings

    Why it is in this group

    • The name "get_top_ten" says "get": a word for reading.
    Description, as the server wrote it
    The ten best services in one capability category today. PRICED: $0.02 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Ranked by a score out of 100 built only from what we measured: liveness history over the last 90 probes, on-chain payment evidence, how many independent registries list it, how many protocols it genuinely answers on, and how completely it describes itself. Frozen nightly after the crawl, so every buyer gets the same list for the same day. A service is placed in a category by the names of the tools it publishes and appears in at most three, so the ten lists are genuinely different from each other.
  • get_trust_scoreMeasured-signal trust score for one endpoint.No findings

    Why it is in this group

    • The name "get_trust_score" says "get": a word for reading.
    Description, as the server wrote it
    Measured-signal trust score for one endpoint. PRICED: $0.02 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Scores an endpoint on liveness history, on-chain payment evidence, cross-registry corroboration and protocol richness. Every input is independently measured by us; no self-reported field contributes.
  • get_weekly_changesWhat came alive and what went dark across the whole index in the last seven days.No findings

    Why it is in this group

    • The name "get_weekly_changes" says "get": a word for reading.
    Description, as the server wrote it
    What came alive and what went dark across the whole index in the last seven days. PRICED: $0.05 in USDC on base via x402 -- no account, no API key, no invoice. Call without `payment` to receive the exact payment requirements, then call again with `payment` to be served. Services that started answering and services that stopped, with index-wide totals, over a window you choose up to 30 days. Rebuilt hourly and served identically to every buyer, so it costs the same to read whether you are first or hundredth.
  • search_agentsSearch the AgentIndex for AI agents, MCP servers, A2A agents, and x402-gated APIs by capability keyword.No findings

    Why it is in this group

    • The name "search_agents" says "search": a word for reading.
    Description, as the server wrote it
    Search the AgentIndex for AI agents, MCP servers, A2A agents, and x402-gated APIs by capability keyword. Returns ranked results with type, description, URL, and tags. FREE for the first 50 calls a day per caller, then $0.005 in USDC on base via x402. Over the allowance, call without `payment` to receive the exact payment requirements, then call again with `payment` to be served.

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 outbound.

Reads private data1
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write1
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.

Tools listed without sign-in

Looked at Nov 20, 2026, 19:03 UTC, unchanged since Nov 13, 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
10
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 10 tools that change things (1 irreversible, 9 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://agents.traderszone.net/mcp: tools/list returned 22 tools to an anonymous client, including payment_evidence (irreversible), preflight_check (write), bulk_score (write), price_check (write), cheapest_that_works (write) and 5 more

    What this check looks for

Posture changes

  1. Nov 13, 2026
    • Reachable: no to yes
    • HTTP status of an anonymous initialize: none to 200
    • tools/list served without credentials: no to yes
    • Tools that change things, listed without credentials: 0 to 10
    • Protocol version: none to 2025-06-18
  2. Nov 6, 2026
    • Reachable: yes to no
    • HTTP status of an anonymous initialize: 200 to none
    • tools/list served without credentials: yes to no
    • Tools that change things, listed without credentials: 10 to 0
    • Protocol version: 2025-06-18 to none

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.

  1. PostureRemote posture changed. Reachable: no -> yes; HTTP status of an anonymous initialize: none -> 200; tools/list served without credentials: no -> yes (and 2 more).View diff
    • Reachable: no to yes
    • HTTP status of an anonymous initialize: none to 200
    • tools/list served without credentials: no to yes
    • Tools that change things, listed without credentials: 0 to 10
    • Protocol version: none to 2025-06-18
  2. PostureRemote posture changed. Reachable: yes -> no; HTTP status of an anonymous initialize: 200 -> none; tools/list served without credentials: yes -> no (and 2 more).View diff
    • Reachable: yes to no
    • HTTP status of an anonymous initialize: 200 to none
    • tools/list served without credentials: yes to no
    • Tools that change things, listed without credentials: 10 to 0
    • Protocol version: 2025-06-18 to none

Install

Pinned to version 1.1.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http agent-index https://agents.traderszone.net/mcp
claude mcp add --transport http agent-index https://agents.traderszone.net/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__agent-index__search_agents",
      "mcp__agent-index__get_agent_profile",
      "mcp__agent-index__get_demand_preview",
      "mcp__agent-index__find_tool",
      "mcp__agent-index__get_liveness_history",
      "mcp__agent-index__get_trust_score",
      "mcp__agent-index__get_endpoint_alerts",
      "mcp__agent-index__get_demand_report",
      "mcp__agent-index__get_service_report",
      "mcp__agent-index__get_top_ten",
      "mcp__agent-index__get_rankings",
      "mcp__agent-index__get_weekly_changes"
    ],
    "ask": [
      "mcp__agent-index__preflight_check",
      "mcp__agent-index__bulk_score",
      "mcp__agent-index__price_check",
      "mcp__agent-index__cheapest_that_works",
      "mcp__agent-index__new_services",
      "mcp__agent-index__gone_dark",
      "mcp__agent-index__has_it_changed",
      "mcp__agent-index__uptime_history",
      "mcp__agent-index__payment_evidence",
      "mcp__agent-index__bulk_export"
    ],
    "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-index. Review it before you use it.

Capability tags behind 2 rules
  • mcp__agent-index__get_agent_profile untrusted-input
  • mcp__agent-index__get_rankings private-data

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.cengit1/agent-index/verify?hash=a2a641213d65d82620d491593618f53395bbd2a40d6a76fbf2376fe534a86142" | jq -e '.changed == false'

Pinned hash a2a641213d65, from Nov 16, 2026, 13:17 UTC.

Watch

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

https://radar.anlyon.com/servers/io.github.cengit1/agent-index/changes.xml

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

Badge

Anlyon Radar badge for io.github.cengit1/agent-index
[![Anlyon Radar](https://radar.anlyon.com/badge/io.github.cengit1/agent-index.svg)](https://radar.anlyon.com/servers/io.github.cengit1/agent-index)

Source: MCP Registry · repository. Synced Oct 21, 2026, 18:26 UTC.

AgentIndex — measured reliability for agent services MCP server: questions and answers

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

Is the AgentIndex — measured reliability for agent services MCP server safe?

No scan can promise that. What its definitions show: 22 tools, 10 of which can change, send or delete something, and 0 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 AgentIndex — measured reliability for agent services MCP server delete data?

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

Can the AgentIndex — measured reliability for agent services MCP server do things that cannot be undone?

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

What tools does the AgentIndex — measured reliability for agent services MCP server have?

22 tools: 1 irreversible, 9 write, 12 read. payment_evidence, preflight_check, bulk_score, price_check, cheapest_that_works, new_services, gone_dark, has_it_changed and 14 more.

Does the AgentIndex — measured reliability for agent services MCP server need sign-in or OAuth?

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

Does the AgentIndex — measured reliability for agent services MCP server send data to outside services?

None of its tools sends, posts or publishes, or takes a recipient, channel or URL.

Has the AgentIndex — measured reliability for agent services MCP server changed its tools?

Not since Anlyon Radar started tracking it. Last scanned Nov 16, 2026.