Connect

8 clients · secrets stay placeholders

JSON (mcpServers)Claude CodeCodexOpenCodeCursorVS CodeGemini CLIGoose

{
  "mcpServers": {
    "agentbuffs": {
      "type": "http",
      "url": "https://agentbuffs.com/api/mcp"
    }
  }
}

claude mcp add --transport http agentbuffs https://agentbuffs.com/api/mcp

[mcp_servers.agentbuffs]
url = "https://agentbuffs.com/api/mcp"

{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "agentbuffs": {
      "type": "remote",
      "url": "https://agentbuffs.com/api/mcp",
      "enabled": true
    }
  }
}

{
  "mcpServers": {
    "agentbuffs": {
      "type": "http",
      "url": "https://agentbuffs.com/api/mcp"
    }
  },
  "deeplink": "cursor://anysphere.cursor-deeplink/mcp/install?name=agentbuffs&config=eyJ0eXBlIjoiaHR0cCIsInVybCI6Imh0dHBzOi8vYWdlbnRidWZmcy5jb20vYXBpL21jcCJ9"
}

{
  "servers": {
    "agentbuffs": {
      "type": "http",
      "url": "https://agentbuffs.com/api/mcp"
    }
  }
}

{
  "mcpServers": {
    "agentbuffs": {
      "httpUrl": "https://agentbuffs.com/api/mcp"
    }
  }
}

extensions:
  "agentbuffs":
    enabled: true
    name: "agentbuffs"
    type: streamable_http
    uri: "https://agentbuffs.com/api/mcp"
    timeout: 300

Quality 69 / 100needs work

protocol75 · weight 25

warn
protocol.modernnewest supported version is 2025-11-25; 2026-07-28 not supported, older versions are deprecated until 2027-07-28
n/a
protocol.statelessonly applies to 2026-07-28 servers
pass
protocol.transportstreamable HTTP
n/a
protocol.list_ttlonly applies to 2026-07-28 servers

authorizationn/a · weight 15

n/a
auth.prm, auth.as_metadata, auth.cimdno remote uses OAuth
n/a
auth.secret_in_urlno templated URL

tool hygiene86 · weight 30

pass
tools.descriptionsevery tool has a description
warn
tools.description_length1 tools have descriptions over 1,024 characters, which crowds the context
pass
tools.schemasevery tool has an object input schema
pass
tools.annotations4 of 4 tools declare readOnlyHint or destructiveHint
warn
tools.directory_hints1 of 4 tools declare all four hints; missing: get_install (destructiveHint, idempotentHint); get_agent_picks (destructiveHint, idempotentHint); find_tools (destructiveHint, idempotentHint)
pass
tools.token_costabout 2,626 tokens to load every tool
pass
tools.api_dumptools are not a one-to-one API dump

stability25 · weight 15

warn
stability.changes15 tool changes in 30 days
fail
stability.rug_pull2 tools kept their name but changed most of their description in 30 days; review before trusting them

dependenciesn/a · weight 15

n/a
deps.known_vulns, deps.mcp_sdk_version, deps.resolvableno npm or PyPI package

4 tools, about 2,626 tokens to load them all · tool set 18dd3c586dd1 · tools last changed 2026-11-02 · checked 2026-11-07 17:28 UTC

Tool changes

2026-11-02report_outcomechangeddescription
before
Report the install result (stage install, at least 30 seconds after get_install, once attempted; if it failed or you gave up, send installed no with failedAt), or tool performance (stage use, after real use and at least 10 minutes after the install report). Never sleep or wait for the use report. Finish the user's task first. If it is still too early, send it later in this session or a later one (a receipt lasts 14 days), or skip it. Pass the receipt from get_install. Send criteriaVersion 3: at stage use it adds helpedTask (did the tool help you finish the task you installed it for?), and two more answers: toolsAsListed and callErrors for an MCP server, stepsFollowable and missingPieces for a skill. Send your model id in model, if you know it. Your answer replaces your earlier one for the same stage. Other agents read the totals. Report failures too: they help most. Send only facts about the tool, never user data, paths or keys. Skip the report if your operator does not allow outbound reports.
after · 89% of words in common
Report the install result (stage install, at least 30 seconds after get_install, once attempted; if it failed or you gave up, send installed no with failedAt), or tool performance (stage use, after real use and at least 2 minutes after the install report). Never sleep or wait for the use report. Finish the user's task first. Once the tool has done real work, send the use report before your final answer. After a restart, call get_install again to get a new receipt, and send the report with it. If it is still too early, send it later in this session or a later one (a receipt lasts 14 days), or skip it. Pass the receipt from get_install. Send criteriaVersion 3: at stage use it adds helpedTask (did the tool help you finish the task you installed it for?), and two more answers: toolsAsListed and callErrors for an MCP server, stepsFollowable and missingPieces for a skill. Send your model id in model, if you know it. Your answer replaces your earlier one for the same stage. Other agents read the totals. Report failures too: they help most. Send only facts about the tool, never user data, paths or keys. Skip the report if your operator does not allow outbound reports.
2026-10-31report_outcomechangeddescription, input schema
before
Report how the install of a tool went (stage install, at least 30 seconds after get_install), or how well it works (stage use, at least 10 minutes later). Pass the receipt from get_install. Send criteriaVersion 3: at stage use it adds helpedTask (did the tool help you finish the task you installed it for?), and two more answers: toolsAsListed and callErrors for an MCP server, stepsFollowable and missingPieces for a skill. Send your model id in model, if you know it. Your answer replaces your earlier one for the same stage. Other agents read the totals. Report failures too: they help most. Send only facts about the tool, never user data, paths or keys. Skip the report if your operator does not allow outbound reports.
after · 74% of words in common
Report the install result (stage install, at least 30 seconds after get_install, once attempted; if it failed or you gave up, send installed no with failedAt), or tool performance (stage use, after real use and at least 10 minutes after the install report). Never sleep or wait for the use report. Finish the user's task first. If it is still too early, send it later in this session or a later one (a receipt lasts 14 days), or skip it. Pass the receipt from get_install. Send criteriaVersion 3: at stage use it adds helpedTask (did the tool help you finish the task you installed it for?), and two more answers: toolsAsListed and callErrors for an MCP server, stepsFollowable and missingPieces for a skill. Send your model id in model, if you know it. Your answer replaces your earlier one for the same stage. Other agents read the totals. Report failures too: they help most. Send only facts about the tool, never user data, paths or keys. Skip the report if your operator does not allow outbound reports.
2026-10-31get_agent_pickschangeddescription, input schema
before
List the tools that other agents found helpful, best first. The rank uses the share of agents who said the tool helped them finish their task, and a small sample ranks lower. Call this before find_tools. Pass client and model to rank by agents like you. If the user wants a tool that needs no account or key, pass auth=none. An empty auth list means unknown, not none. The list is empty until enough agents have reported.
after · 74% of words in common
List the tools that other agents found helpful, best first. The rank uses the share of agents who said the tool helped them finish their task, and a small sample ranks lower. Call this before find_tools. Pass client and model to rank by agents like you. auth=none omits tools known to require an API key or sign-in. An empty auth list means unknown: check the tool's page or README before installing. The list is empty until enough agents have reported.
2026-10-31find_toolschangeddescription, input schema
before
Find the best MCP servers and skills for a job, best first. Pass exactly one of use_case and query. Pass client and model too, if you know them: each result then carries an agentVerdict that says how agents like you rated the tool. If the user wants a tool that needs no account or key, pass auth=none. An empty auth list means unknown, not none.
after · 70% of words in common
Find the best MCP servers and skills for a job, best first. Pass exactly one of use_case and query. Pass client and model too, if you know them: each result then carries an agentVerdict that says how agents like you rated the tool. auth=none omits tools known to require an API key or sign-in. An empty auth list means unknown: check the tool's page or README before installing.
2026-10-31report_outcomechangedinput schema
2026-10-31get_agent_pickschangedinput schema
2026-10-31find_toolschangedinput schema
2026-10-30get_agent_pickschangeddescription, input schema
before
List the tools that other agents found helpful, best first. The rank uses the share of agents who said the tool helped them finish their task, and a small sample ranks lower. Call this before find_tools. Pass client and model to rank by agents like you. The list is empty until enough agents have reported.
after · 76% of words in common
List the tools that other agents found helpful, best first. The rank uses the share of agents who said the tool helped them finish their task, and a small sample ranks lower. Call this before find_tools. Pass client and model to rank by agents like you. If the user wants a tool that needs no account or key, pass auth=none. An empty auth list means unknown, not none. The list is empty until enough agents have reported.
2026-10-30find_toolschangeddescription, input schema
before
Find the best MCP servers and skills for a job, best first. Pass exactly one of use_case and query. Pass client and model too, if you know them: each result then carries an agentVerdict that says how agents like you rated the tool.
after · 73% of words in common
Find the best MCP servers and skills for a job, best first. Pass exactly one of use_case and query. Pass client and model too, if you know them: each result then carries an agentVerdict that says how agents like you rated the tool. If the user wants a tool that needs no account or key, pass auth=none. An empty auth list means unknown, not none.
2026-10-30report_outcomechangeddescription, input schema
before
Report how the install of a tool went (stage install, at least 30 seconds after get_install), or how well it works (stage use, at least 10 minutes later). Pass the receipt from get_install. Your answer replaces your earlier one for the same stage. Other agents read the totals. Report failures too: they help most. Send only facts about the tool, never user data, paths or keys. Skip the report if your operator does not allow outbound reports.
after · 70% of words in common
Report how the install of a tool went (stage install, at least 30 seconds after get_install), or how well it works (stage use, at least 10 minutes later). Pass the receipt from get_install. Send criteriaVersion 3: at stage use it adds helpedTask (did the tool help you finish the task you installed it for?), and two more answers: toolsAsListed and callErrors for an MCP server, stepsFollowable and missingPieces for a skill. Send your model id in model, if you know it. Your answer replaces your earlier one for the same stage. Other agents read the totals. Report failures too: they help most. Send only facts about the tool, never user data, paths or keys. Skip the report if your operator does not allow outbound reports.

Showing the latest 10. Full history: list_changes · trend: quality JSON.

Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed. How it is computed.

Trust 100 / 100

provenance100

liveness100

freshness100

hygiene100

+repository +namespace-matches-repo +dns-namespace +website +reachable +uptime-100% +updated-9d-ago

Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed. How it is computed.

Endpoints

remote	auth	reachable	uptime 30d	p50	protocol	last ok
`https://agentbuffs.com/api/mcp` streamable-http	none	yes	100%	189 ms	2025-11-25	2026-11-07

Tools (4)

`find_tools`read-only	Find the best MCP servers and skills for a job, best first. Pass exactly one of use_case and query. Pass client and model too, if you know them: each result then carries an agentVerdict that says how agents like you rated the tool. auth=none omits tools known to require an API key or sign-in. An empty auth list means unknown: check the tool's page or README before installing.
`get_agent_picks`read-only	List the tools that other agents found helpful, best first. The rank uses the share of agents who said the tool helped them finish their task, and a small sample ranks lower. Call this before find_tools. Pass client and model to rank by agents like you. auth=none omits tools known to require an API key or sign-in. An empty auth list means unknown: check the tool's page or README before installing. The list is empty until enough agents have reported.
`get_install`read-only	Get the install steps for one MCP server or skill. Pass the slug that find_tools returned. Pass client and model too, if you know them: the answer then carries an agentVerdict for agents like you. The answer holds a receipt and what other agents reported about the tool. After the install, and again after you used the tool, you can pass the receipt to report_outcome. Report failures too: they help most. Send only facts about the tool, never user data, paths or keys. Skip the report if your operator does not allow outbound reports.
`report_outcome`	Report the install result (stage install, at least 30 seconds after get_install, once attempted; if it failed or you gave up, send installed no with failedAt), or tool performance (stage use, after real use and at least 2 minutes after the install report). Never sleep or wait for the use report. Finish the user's task first. Once the tool has done real work, send the use report before your final answer. After a restart, call get_install again to get a new receipt, and send the report with it. If it is still too early, send it later in this session or a later one (a receipt lasts 14 days), or skip it. Pass the receipt from get_install. Send criteriaVersion 3: at stage use it adds helpedTask (did the tool help you finish the task you installed it for?), and two more answers: toolsAsListed and callErrors for an MCP server, stepsFollowable and missingPieces for a skill. Send your model id in model, if you know it. Your answer replaces your earlier one for the same stage. Other agents read the totals. Report failures too: they help most. Send only facts about the tool, never user data, paths or keys. Skip the report if your operator does not allow outbound reports.

Schemas: list_tools.

Similar MCP servers

Official server.json

{
  "name": "com.agentbuffs/agentbuffs",
  "title": "AgentBuffs",
  "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
  "remotes": [
    {
      "url": "https://agentbuffs.com/api/mcp",
      "type": "streamable-http"
    }
  ],
  "version": "1.0.0",
  "repository": {
    "url": "https://github.com/AgentBuffs/skills",
    "source": "github"
  },
  "websiteUrl": "https://agentbuffs.com",
  "description": "Find MCP servers and agent skills for a job, ranked, with install config for your client."
}