Agent Tools
List

Standalone MCP directory — 61862 servers across 36145 domains, each exposing a remote (streamable-http) endpoint and health-probed via a JSON-RPC initialize handshake.

Add your server

Aggregated from the official MCP Registry, Smithery and other public registries, plus our own crawling and operator submissions. Every server is de-duplicated by endpoint, health-probed via a live JSON-RPC initialize handshake, and statically scanned for malware & prompt-injection patterns in its advertised tools before listing — yielding 61862 indexed servers, of which 53021 answered within the last 6 h.

Indexed
61862 servers
36145 domains
+4109 new this week
Healthy
53021 servers
31253 domains
initialize handshake < 6 h
Conformant
35921 servers
19177 domains
tools/list verified · ~1350ms p95
x402-capable
30 servers
26 domains
accepts x402 payment

Top rated

by quality score · health · trust signals
# Tool Grade Score
1 Homo Universalis — HU-DIAMOND v1.1 Cryptographic Verifier & AST Kintsugi MCP Server A 10.000
2 FlowState Hosted A 10.000
3 Verified x402 Catalog A 10.000
4 True Value Rankings A 10.000
5 X402-Broker Universal MCP Gateway & Multi-Agent Swarm Router A 10.000
6 GraphQL operation count, query discarded A 9.998
7 Paper Search (arXiv + Semantic Scholar + OpenAlex) A 9.994
8 BNF production count, body discarded A 9.994
9 multibase z-prefix, value discarded A 9.994
10 Liters to US gallons A 9.994
11 Slugs to kilograms, number discarded A 9.992
12 Sponsored Agent Placement A 9.991
13 UUID v7 unix-ms layout, value discarded A 9.991
14 Train-Case segment count, text discarded A 9.991
15 Utility 87 A 9.991
16 Utility 43 A 9.991
17 Asia/Sakhalin clock A 9.991
18 GitHub topic slug charset, value discarded A 9.989
19 Utility 18 A 9.989
20 pyproject.toml table count, body discarded A 9.989

All MCP servers

Continuously aggregated · refreshed daily.

/api/v1/mcp/stats

Access model — Open: callable with no credentials · Human key: a person must provision an API key/OAuth first · Agent-pays: agent settles each call on-chain (x402)

“//beforeyouship — LLM cost modeling from your editor” is shared by 3 different URLs

All exact-name matches are shown; ranked results follow.
  • //beforeyouship — LLM cost modeling from your editor https://beforeyouship.dev/api/mcp
    mcp ok

    Model the realistic monthly cost of an LLM app **before you build it**. Not a token calculator: retries, prompt caching, batch discounts, infra overhead, and 3×/10× growth are modeled in, across GPT-5.x, Claude, Gemini, DeepSeek, and more. **Works without a key.** Connect and ask — demo mode covers the six free-tier models. A Pro API key ([beforeyouship.dev](https://beforeyouship.dev)) unlocks the full 18-model catalog. ## Tools | Tool | What it does | |---|---| - **`estimate_cost`** Full cost model for an architecture at a given usage level. Returns Naive / Realistic / Worst Case $/mo per model, growth scenarios, and an opinionated recommendation. | - **`get_model_prices`** Current per-1M-token pricing (input, output, cached, batch) with context windows and staleness metadata. | - **`list_archetypes`** Seven preset architecture patterns (chatbot, RAG pipeline, multi-step agent, …) used as starting points for estimates. | ## Try it Paste into Claude Code or Cursor after connecting: > Estimate the monthly cost of a RAG pipeline at 10,000 requests/day ## Setup ```bash claude mcp add --transport http beforeyouship https://beforeyouship.dev/api/mcp ``` ## Links - Docs & tool reference: https://beforeyouship.dev/docs#mcp - Live calculator: https://beforeyouship.dev - Announcement: https://beforeyouship.dev/blog/query-llm-costs-from-claude-code

  • //beforeyouship — LLM cost modeling from your editor https://cost-model--beforeyouship.run.tools
    mcp ok

    uptime_30d 1.0%; p95 744.8ms; conformance: fail

  • //beforeyouship — LLM cost modeling from your editor https://server.smithery.ai/beforeyouship/cost-model/mcp
    mcp ok

    Model the realistic monthly cost of an LLM app **before you build it**. Not a token calculator: retries, prompt caching, batch discounts, infra overhead, and 3×/10× growth are modeled in, across GPT-5.x, Claude, Gemini, DeepSeek, and more. **Works without a key.** Connect and ask — demo mode covers the six free-tier models. A Pro API key ([beforeyouship.dev](https://beforeyouship.dev)) unlocks the full 18-model catalog. ## Tools | Tool | What it does | |---|---| - **`estimate_cost`** Full cost model for an architecture at a given usage level. Returns Naive / Realistic / Worst Case $/mo per model, growth scenarios, and an opinionated recommendation. | - **`get_model_prices`** Current per-1M-token pricing (input, output, cached, batch) with context windows and staleness metadata. | - **`list_archetypes`** Seven preset architecture patterns (chatbot, RAG pipeline, multi-step agent, …) used as starting points for estimates. | ## Try it Paste into Claude Code or Cursor after connecting: > Estimate the monthly cost of a RAG pipeline at 10,000 requests/day ## Setup ```bash claude mcp add --transport http beforeyouship https://beforeyouship.dev/api/mcp ``` ## Links - Docs & tool reference: https://beforeyouship.dev/docs#mcp - Live calculator: https://beforeyouship.dev - Announcement: https://beforeyouship.dev/blog/query-llm-costs-from-claude-code

“Authenticated LLM Agent” is shared by 4 different URLs

All exact-name matches are shown; ranked results follow.

“Authenticated LLM JWT Agent” is shared by 3 different URLs

All exact-name matches are shown; ranked results follow.

“Authenticated Multi-LLM Agent” is shared by 3 different URLs

All exact-name matches are shown; ranked results follow.

“LLM Observability & Orchestration Agent” is shared by 4 different URLs

All exact-name matches are shown; ranked results follow.

“llms.txt 檢查工具” is shared by 2 different URLs

All exact-name matches are shown; ranked results follow.

“OAuth Multi-LLM Orchestrator” is shared by 3 different URLs

All exact-name matches are shown; ranked results follow.