io.github.cyanheads/cyanheads-mcp-server
Fleet discovery for the cyanheads MCP ecosystem — semantic search + install snippets.
Que pouvez-vous faire avec io.github.cyanheads/cyanheads-mcp-server ?
- Hourly background catalog refresh (
CATALOG_REFRESH_SECONDS, default3600) with an atomic index swap whengeneratedAtchanges — no restart needed - Empty catalogs, conflicting duplicate identities, and vectors that can't be normalized (wrong dimension, all zeros, or a non-finite component) fail before index replacement; refresh retains the last valid catalog. Identical duplicates collapse before indexi...
- Query embeddings are computed at request time via
@huggingface/transformers; document embeddings are pre-computed, L2-normalized, and Matryoshka-truncated, shipped insidefleet.json. The server normalizes each document vector again at load, so scores st... - The embedding model is warmed up during startup, before OpenTelemetry's HTTP instrumentation patches
fetch— avoids a cold-cache model-load failure under OTEL - Self-describing:
cyanheads_describe_entryresolves this server's own name and tools from a static fallback record, consulted only when the remote catalog doesn't carry an entry for it CATALOG_URLcan point at any endpoint serving the same schema, to front a custom fleet- Search responses echo the effective query, report the total match count before the limit, include broadening guidance when nothing matches, and name the offset to continue from when a
limitabove 20 is cut to the per-call cap - Discriminated
result.kind(tool|server) oncyanheads_describe_entrylets callers branch on data, not string parsing
Installation
Ajoutez ceci à la configuration MCP de votre client (Claude Desktop : claude_desktop_config.json ; Cursor : .cursor/mcp.json ; VS Code et les autres utilisent la même forme « mcpServers »).
claude mcp add --transport http cyanheads-mcp-server 'https://cyanheads.caseyjhand.com/mcp'
{
"mcpServers": {
"cyanheads-mcp-server": {
"type": "http",
"url": "https://cyanheads.caseyjhand.com/mcp"
}
}
}
@cyanheads/cyanheads-mcp-serverclaude mcp add cyanheads-mcp-server -e 'MCP_LOG_LEVEL=info' -e 'CATALOG_URL=https://caseyjhand.com/fleet.json' -e 'CATALOG_FETCH_TIMEOUT_MS=10000' -e 'CATALOG_REFRESH_SECONDS=3600' -e 'EMBEDDING_MODEL_ID=Snowflake/snowflake-arctic-embed-m-v1.5' -e 'SIMILARITY_FLOOR=0.3' -- npx -y @cyanheads/cyanheads-mcp-server@0.5.0
{
"mcpServers": {
"cyanheads-mcp-server": {
"command": "npx",
"args": [
"-y",
"@cyanheads/cyanheads-mcp-server@0.5.0"
],
"env": {
"MCP_LOG_LEVEL": "info",
"CATALOG_URL": "https://caseyjhand.com/fleet.json",
"CATALOG_FETCH_TIMEOUT_MS": "10000",
"CATALOG_REFRESH_SECONDS": "3600",
"EMBEDDING_MODEL_ID": "Snowflake/snowflake-arctic-embed-m-v1.5",
"SIMILARITY_FLOOR": "0.3"
}
}
}
}
Variables d'environnement
MCP_LOG_LEVELMinimum log level: debug, info, notice, warning, error.CATALOG_URLRemote fleet.json endpoint (schema v2 with baked embeddings).CATALOG_FETCH_TIMEOUT_MSPer-request timeout for fleet.json fetches: integer milliseconds from 1 to 2147483647.CATALOG_REFRESH_SECONDSBackground poll interval for fleet.json refreshes in seconds. 0 disables refresh; positive values up to 2147483.647 use the runtime's millisecond timer resolution.EMBEDDING_MODEL_IDHugging Face model id for query embedding. Must match fleet.json.embeddingModel.SIMILARITY_FLOORCosine similarity cutoff for cyanheads_search_catalog results. Must be within [0, 1].
@cyanheads/cyanheads-mcp-serverclaude mcp add cyanheads-mcp-server -e 'MCP_TRANSPORT_TYPE=<MCP_TRANSPORT_TYPE>' -e 'MCP_HTTP_HOST=127.0.0.1' -e 'MCP_HTTP_PORT=3010' -e 'MCP_HTTP_ENDPOINT_PATH=/mcp' -e 'MCP_AUTH_MODE=none' -e 'MCP_LOG_LEVEL=info' -e 'CATALOG_URL=https://caseyjhand.com/fleet.json' -e 'CATALOG_FETCH_TIMEOUT_MS=10000' -e 'CATALOG_REFRESH_SECONDS=3600' -e 'EMBEDDING_MODEL_ID=Snowflake/snowflake-arctic-embed-m-v1.5' -e 'SIMILARITY_FLOOR=0.3' -- npx -y @cyanheads/cyanheads-mcp-server@0.5.0
{
"mcpServers": {
"cyanheads-mcp-server": {
"command": "npx",
"args": [
"-y",
"@cyanheads/cyanheads-mcp-server@0.5.0"
],
"env": {
"MCP_TRANSPORT_TYPE": "<MCP_TRANSPORT_TYPE>",
"MCP_HTTP_HOST": "127.0.0.1",
"MCP_HTTP_PORT": "3010",
"MCP_HTTP_ENDPOINT_PATH": "/mcp",
"MCP_AUTH_MODE": "none",
"MCP_LOG_LEVEL": "info",
"CATALOG_URL": "https://caseyjhand.com/fleet.json",
"CATALOG_FETCH_TIMEOUT_MS": "10000",
"CATALOG_REFRESH_SECONDS": "3600",
"EMBEDDING_MODEL_ID": "Snowflake/snowflake-arctic-embed-m-v1.5",
"SIMILARITY_FLOOR": "0.3"
}
}
}
}
Variables d'environnement
MCP_TRANSPORT_TYPESelects the HTTP transport.MCP_HTTP_HOSTHostname for the HTTP server.MCP_HTTP_PORTPort to run the HTTP server on.MCP_HTTP_ENDPOINT_PATHEndpoint path for the MCP server.MCP_AUTH_MODEAuthentication mode: none, jwt, or oauth.MCP_LOG_LEVELMinimum log level: debug, info, notice, warning, error.CATALOG_URLRemote fleet.json endpoint (schema v2 with baked embeddings).CATALOG_FETCH_TIMEOUT_MSPer-request timeout for fleet.json fetches: integer milliseconds from 1 to 2147483647.CATALOG_REFRESH_SECONDSBackground poll interval for fleet.json refreshes in seconds. 0 disables refresh; positive values up to 2147483.647 use the runtime's millisecond timer resolution.EMBEDDING_MODEL_IDHugging Face model id for query embedding. Must match fleet.json.embeddingModel.SIMILARITY_FLOORCosine similarity cutoff for cyanheads_search_catalog results. Must be within [0, 1].
Documentation
Tirée du README du projet sur GitHub, actualisée chaque semaine. Les liens ouvrent GitHub.
Sommaire
@cyanheads/cyanheads-mcp-server
Fleet discovery for the cyanheads MCP ecosystem — semantic search + install snippets.
2 Tools • 0 Resources • 0 Prompts
Public Hosted Server: https://cyanheads.caseyjhand.com/mcp
Overview
Find the right MCP tool for the job across the cyanheads fleet. Describe a task in plain language to search every tool and server, then get a description, connection URL, and install snippet for your client.
Tools
| Tool | Description |
|---|---|
cyanheads_search_catalog | Search fleet tools and servers by natural-language query. Returns ranked matches with brief summaries and the owning server. |
cyanheads_describe_entry | Return the description, connection URL, and per-client install snippet for a named tool or server. |
Capability reference
cyanheads_search_catalog tool
query1–500 raw characters, with non-whitespace text required; surrounding whitespace is trimmed before embedding and echo.scopeselectstools(default) orservers; optionalcategory:research,government,public-data,utility;limit1–1000 (default 5), served at most 20 per call: when a largerlimitleaves matches past the page, the response addstruncated,shown,cap, and anoticenaming the offset to continue from.- Results carry cosine-similarity
score([0, 1], comparable within one response). Tool searches include aserversroll-up with ten entries per page andserversTotal; an unloaded catalog returns retryablecatalog_empty. - Page results with
offsetand the tools-scope roll-up independently withserversOffset(both default to 0). Reuse the same query, scope, category, and limit withnextOffset/nextServersOffset; null means exhausted. Each call uses the current catalog, so a refresh between pages can change ordering. Servers scope requiresserversOffset: 0and omits roll-up fields. SIMILARITY_FLOOR(default0.3) filters matches before the limit applies.- An empty
categorystring is treated as omitted; other unknown category values are rejected.
cyanheads_describe_entry tool
name1–64 characters accepts a tool or server name, auto-detected or pinned viakind. Exact names win, then unique case-insensitive names; output uses canonical names.clientoptionally selectsclaude-code,cline,codex,cursor,gemini,streamable-http(genericmcpServersJSON), orcurl.server(optional, 1–64 characters) names the owning server: a search result'sname+serverdescribes exactly that row, in either scope. It is required when more than one server owns the tool name; with a server name, it must name the same server.- Empty
kind,client, andserverstrings are treated as omitted. Other unknownkindorclientvalues are rejected. - Both kinds return connection metadata and install snippets; server results also include the full tool inventory. Published servers get local snippets; hosted servers also get HTTP snippets. For local-only entries,
client: "curl"returns successful metadata with aninstallNoticenaming a stdio client to try. Required environment variables remain visible even without snippets. - Descriptions are catalog text; obtain tool schemas from the connected server's
tools/list. - Failures return
not_found,ambiguous_kind(setkind),ambiguous_owner(setserver; the error lists the owners), or retryablecatalog_empty.
Features
Built on @cyanheads/mcp-ts-core: stdio and Streamable HTTP transports, pluggable auth (none / jwt / oauth), swappable storage (in-memory, filesystem, Supabase, Cloudflare KV/R2/D1), structured logging with optional OpenTelemetry tracing.
Catalog-specific:
- Hourly background catalog refresh (
CATALOG_REFRESH_SECONDS, default3600) with an atomic index swap whengeneratedAtchanges — no restart needed - Empty catalogs, conflicting duplicate identities, and vectors that can't be normalized (wrong dimension, all zeros, or a non-finite component) fail before index replacement; refresh retains the last valid catalog. Identical duplicates collapse before indexing, with a bounded load-time warning; servers with no tools remain valid.
- Query embeddings are computed at request time via
@huggingface/transformers; document embeddings are pre-computed, L2-normalized, and Matryoshka-truncated, shipped insidefleet.json. The server normalizes each document vector again at load, so scores stay cosine similarities even when a catalog's vectors drift from unit length, and a bounded warning names vectors more than1e-4off. - The embedding model is warmed up during startup, before OpenTelemetry's HTTP instrumentation patches
fetch— avoids a cold-cache model-load failure under OTEL - Self-describing:
cyanheads_describe_entryresolves this server's own name and tools from a static fallback record, consulted only when the remote catalog doesn't carry an entry for it CATALOG_URLcan point at any endpoint serving the same schema, to front a custom fleet
Agent-friendly output:
- Search responses echo the effective query, report the total match count before the limit, include broadening guidance when nothing matches, and name the offset to continue from when a
limitabove 20 is cut to the per-call cap - Discriminated
result.kind(tool|server) oncyanheads_describe_entrylets callers branch on data, not string parsing - Every search result carries a
scorefield for trust calibration, plus aserversroll-up so agents can see which servers matched without a second call
Getting started
Public Hosted Instance
A public instance is available at https://cyanheads.caseyjhand.com/mcp — no installation required. It serves Streamable HTTP; each client takes its own config shape.
For Claude Code:
claude mcp add --transport http cyanheads https://cyanheads.caseyjhand.com/mcp
For Cline (cline_mcp_settings.json):
{
"mcpServers": {
"cyanheads-mcp-server": {
"type": "streamableHttp",
"url": "https://cyanheads.caseyjhand.com/mcp"
}
}
}
For Cursor (mcp.json):
{
"mcpServers": {
"cyanheads-mcp-server": {
"url": "https://cyanheads.caseyjhand.com/mcp"
}
}
}
Claude Desktop adds remote servers under Settings → Connectors.
Claude Desktop bundle
Download the MCPB bundle for a local stdio install. It includes ONNX WebAssembly inference, so it does not need platform-specific native libraries. The first start downloads the embedding model; subsequent starts reuse the model cache. npm and Docker installs use native CPU inference.
Self-Hosted / Local
Add the following to your MCP client configuration file.
{
"mcpServers": {
"cyanheads-mcp-server": {
"type": "stdio",
"command": "bunx",
"args": ["@cyanheads/cyanheads-mcp-server@latest"],
"env": {
"MCP_TRANSPORT_TYPE": "stdio",
"MCP_LOG_LEVEL": "info"
}
}
}
}
Or with npx (no Bun required):
{
"mcpServers": {
"cyanheads-mcp-server": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@cyanheads/cyanheads-mcp-server@latest"],
"env": {
"MCP_TRANSPORT_TYPE": "stdio",
"MCP_LOG_LEVEL": "info"
}
}
}
}
Or with Docker:
{
"mcpServers": {
"cyanheads-mcp-server": {
"type": "stdio",
"command": "docker",
"args": ["run", "-i", "--rm", "-e", "MCP_TRANSPORT_TYPE=stdio", "ghcr.io/cyanheads/cyanheads-mcp-server:latest"]
}
}
}
For Streamable HTTP, set the transport and start the server:
MCP_TRANSPORT_TYPE=http MCP_HTTP_PORT=3010 bun run start:http
# Server listens at http://localhost:3010/mcp
Prerequisites
- Bun v1.4.0 or higher (or Node.js v24+).
Installation
- Clone the repository:
git clone https://github.com/cyanheads/cyanheads-mcp-server.git
- Navigate into the directory:
cd cyanheads-mcp-server
- Install dependencies:
bun install
- Configure environment:
cp .env.example .env
# edit .env to override any defaults
Configuration
All configuration is validated at startup via Zod schemas in src/config/server-config.ts. Every variable has a sensible default — out of the box, the server points at the canonical cyanheads fleet.
| Variable | Description | Default |
|---|---|---|
MCP_TRANSPORT_TYPE | Transport: stdio or http | stdio |
MCP_HTTP_PORT | HTTP server port | 3010 |
MCP_HTTP_HOST | HTTP server bind host | 127.0.0.1 |
MCP_HTTP_ENDPOINT_PATH | HTTP endpoint path where the MCP server is mounted | /mcp |
MCP_AUTH_MODE | Authentication: none, jwt, or oauth | none |
MCP_SESSION_MODE | HTTP session posture: auto, stateful, or stateless. src/index.ts declares stateless; set this only to override it. | stateless |
MCP_LOG_LEVEL | Log level (debug, info, warning, error, etc.) | info |
CATALOG_URL | Remote fleet.json endpoint (schema v2 with baked embeddings). Must be an absolute URL. Override to front your own fleet. | https://caseyjhand.com/fleet.json |
CATALOG_FETCH_TIMEOUT_MS | Per-request timeout for fleet.json fetches: integer milliseconds, 1–2147483647. | 10000 |
CATALOG_REFRESH_SECONDS | Background poll interval for fleet.json refresh. 0 disables; positive values up to 2147483.647 seconds use the runtime's millisecond timer resolution. | 3600 |
EMBEDDING_MODEL_ID | Hugging Face model id for query embedding. Must match fleet.json.embeddingModel. | Snowflake/snowflake-arctic-embed-m-v1.5 |
SIMILARITY_FLOOR | Cosine similarity cutoff for cyanheads_search_catalog results. Must be within [0, 1]. | 0.3 |
OTEL_ENABLED | Enable OpenTelemetry | false |
OTEL_EXPORTER_OTLP_ENDPOINT | OTLP base URL for traces and metrics | Unset |
OTEL_EXPORTER_OTLP_LOGS_ENDPOINT | Explicit OTLP logs URL; requires the OTel log peers included in the default Docker build | Unset |
LOG_TOOL_FAILURE_PAYLOADS | Log failed arguments and results, redacted by key name. Secrets inside free-form values are not redacted. | false |
LOG_TOOL_FAILURE_PAYLOAD_MAX_BYTES | UTF-8 byte cap for each logged failure payload | 16384 |
See .env.example for the full list of optional overrides.
Running the server
Local development
- Build and run the production version:
# One-time build bun run rebuild # Run the built server bun run start:http # or bun run start:stdio - Run checks and tests:
bun run devcheck # Lints, formats, type-checks, runs MCP and packaging linters bun run test # Runs the test suite
Docker
docker build -t cyanheads-mcp-server .
docker run --rm -p 3010:3010 cyanheads-mcp-server
The Dockerfile defaults to HTTP transport, stateless session mode, and logs to /var/log/cyanheads-mcp-server. OpenTelemetry peer dependencies are installed by default — build with --build-arg OTEL_ENABLED=false to omit them.
Project structure
| Directory | Purpose |
|---|---|
src/mcp-server/tools | Tool definitions (*.tool.ts). Two tools — cyanheads_search_catalog and cyanheads_describe_entry. |
src/services/catalog | Catalog service — remote fleet.json provider with atomic-swap refresh, vector index, snippet builders, the self-description fallback record, and the query-time embedding runtime (@huggingface/transformers, behind an injectable interface for deterministic tests). |
src/config | Server-specific environment variable parsing and validation with Zod. |
tests/ | Unit and integration tests, mirroring the src/ structure. |
docs/ | Design doc and schema reference. |
Development guide
See CLAUDE.md for development guidelines and architectural rules. The short version:
- Handlers throw, framework catches — no
try/catchin tool logic - Use
ctx.logfor logging,ctx.statefor storage - Register new tools in the
toolsarray passed tocreateApp()insrc/index.ts - Wrap external data: validate raw → normalize to domain type → return output schema; never fabricate missing fields
Contributing
Issues are welcome. Run checks and tests before submitting:
bun run devcheck
bun run test
License
Apache-2.0 — see LICENSE for details.
Questions fréquentes
Qu'est-ce que io.github.cyanheads/cyanheads-mcp-server ?
Fleet discovery for the cyanheads MCP ecosystem — semantic search + install snippets.
Comment installer io.github.cyanheads/cyanheads-mcp-server ?
Il fonctionne comme serveur distant. Dans Claude Code : claude mcp add --transport http cyanheads-mcp-server 'https://cyanheads.caseyjhand.com/mcp'. Pour les autres clients, collez la configuration JSON de cette page dans leurs réglages MCP.
Faut-il une clé API pour io.github.cyanheads/cyanheads-mcp-server ?
Son entrée au registre n'indique aucune clé ni secret obligatoire. Certaines fonctions peuvent tout de même demander un compte auprès du service auquel il se connecte.
io.github.cyanheads/cyanheads-mcp-server est-il open source ?
Oui. Son code est sur GitHub sous licence Apache-2.0.
Qui maintient io.github.cyanheads/cyanheads-mcp-server ?
Il est publié par cyanheads. AI.tn le référence sans le maintenir ni le recommander.
Issu du registre MCP officiel. Être listé n'est pas une recommandation : vérifiez le code et les permissions d'un serveur avant de lui donner accès à vos données.