The MCP tool directory.

Tool to generate the TickTick OAuth2 authorization URL. Use to redirect the user to obtain the authorization code (step 1).

Tool to redirect user to TickTick authorization page to obtain authorization code. Use when initiating the OAuth2 flow before exchanging the code.

Tool to update an existing project. Use when you need to modify project details like name, color, sort order, view mode, or kind after selecting a project ID.

Tool to update an existing task. Use after confirming the taskId and projectId.

Create a service token for machine-to-machine authentication. Accepts API key or bearer token. Optionally apply restrictions.

Creates a new session for the tool router feature. This endpoint initializes a new session with specified toolkits and their authentication configurations. The session provides an isolated environment for testing and managing tool routing logic with scoped MCP server access.

Executes a specific tool within a tool router session. This is the primary execution endpoint for both meta tools and app tools exposed by the session. The toolkit is automatically inferred from the tool slug. For app tools, the tool must belong to an allowed toolkit and must not be disabled in the session configuration. The endpoint validates permissions, resolves connected accounts when needed, and executes the tool with the session context. The top-level account field applies only to direct app tool execution in multi-account sessions. Meta/helper tools either ignore it or define their own account-selection fields, for example COMPOSIO_MULTI_EXECUTE_TOOL.tools[].account.

Executes a Composio meta tool (COMPOSIO_*) within a tool router session. This endpoint is kept for meta-tool compatibility; clients can also use the primary /execute endpoint.

Retrieves an existing tool router session by its ID. Returns the session configuration, MCP server URL, and available tools.

Initiates an authentication link session for a specific toolkit within a tool router session. Returns a link token and redirect URL that users can use to complete the OAuth flow.

Deletes a file from a workbench session storage mount. S3 delete is idempotent — deleting a non-existent file succeeds silently.

Generates a presigned download URL for a file in a workbench session mount. Accepts a relative path within the mount.

Lists files in a workbench session storage mount with cursor-based pagination. Use the download_url endpoint with the returned mount_relative_path to get a presigned download URL.

Generates a presigned upload URL for uploading a file to a workbench session mount. The caller should PUT the file content directly to the returned URL.

Execute any native API call on a toolkit with authentication automatically injected from Composio. This endpoint proxies HTTP requests to third-party APIs using connected account credentials resolved from the session context. Provide the toolkit slug, API endpoint, and HTTP method — Composio handles authentication injection, abstracting away credential management. Supports all HTTP methods, custom headers/query parameters, and binary request/response bodies.

Search for tools matching a given use case query within a tool router session. Returns matching tool slugs, full tool schemas, toolkit connection statuses, and workflow guidance in a predictable format.

Retrieves a cursor-paginated list of toolkits available in the tool router session. Includes toolkit metadata, composio-managed auth schemes, and connected accounts if available. Optionally filter by specific toolkit slugs.

Returns the tools available in a tool router session with their complete schemas. This includes both meta tools and any preloaded app tools exposed by the session.

Partially updates the configuration of an existing tool router session. Only the fields provided in the request body will be updated. Uses optimistic concurrency control to prevent lost updates. The previous config is stored in config history.

Retrieves a comprehensive list of all available toolkit categories from their latest versions. These categories can be used to filter toolkits by type or purpose when using the toolkit listing endpoint. Categories help organize toolkits into logical groups based on their functionality or industry focus.

Retrieves the last 10 versions changelog for all toolkits. This endpoint provides version history and changelog information for each toolkit.

Retrieves comprehensive information about a specific toolkit using its unique slug identifier. This endpoint provides detailed metadata, authentication configuration options, and feature counts for the requested toolkit.

Retrieves a comprehensive list of toolkits of their latest versions that are available to the authenticated project. Toolkits represent integration points with external services and applications, each containing a collection of tools and triggers. This endpoint supports filtering by category and management type, as well as different sorting options.

Retrieves a comprehensive list of toolkits of their latest versions that are available to the authenticated project. Toolkits represent integration points with external services and applications, each containing a collection of tools and triggers. This endpoint supports filtering by category and management type, as well as different sorting options. You can optionally specify a list of toolkit slugs to fetch specific toolkits.

Detect whether a product's MCP tool schema has DRIFTED since you cached it — rug-pull / tool-poisoning detection (a server silently changing a tool's description or input schema after you approved it; web search structurally cannot answer this). Pass the product slug and EITHER the schema hash you cached earlier (knownHash, from products.get_mcp_setup → server.schema.hash — the strongest signal: an exact mismatch means the tools changed) OR the ISO-8601 timestamp you cached at (since). With neither, it returns the current fingerprint to cache for next time. A product may expose MULTIPLE servers: knownHash is per-server, so pass `server` (a qualifiedName) with knownHash on a multi-server product — otherwise knownHash is applied only when there's exactly one server. Re-verify before trusting a previously-approved tool. Response: { product, hasMcpSupport, drifted (did ANY tracked server change vs your reference; null when no reference given), servers[] (each: qualifiedName, currentSchemaHash, schemaStable, lastSchemaChangeAt, toolCount, directlyProbed, driftedSinceKnownHash, driftedSince, advice) }. Errors: { error: { code: 'not_found'|'bad_input', ... } }.

Retrieve a list of all available tool enumeration values (tool slugs) from latest version of each toolkit. This endpoint returns a comma-separated string of tool slugs that can be used in other API calls.

Execute a specific tool operation with provided arguments and authentication. This is the primary endpoint for integrating with third-party services and executing tools. You can provide structured arguments or use natural language processing by providing a text description of what you want to accomplish.

Uses AI to translate a natural language description into structured arguments for a specific tool. This endpoint is useful when you want to let users describe what they want to do in plain language instead of providing structured parameters.

Proxy an HTTP request to a third-party API using connected account credentials. This endpoint allows making authenticated API calls to external services while abstracting away authentication details.

Find the best MCP tools for a given task. Describe what you want to accomplish (e.g. 'manage github issues', 'send slack messages', 'query a database') and get ranked results — each result is one (product, tool) pair, since same-named tools across providers are NOT interchangeable. By default only shows tools available via remote (network-hosted) servers. Audience: agent builders looking for installable MCP tools — for B2B SaaS recommendations, prefer products.search or products.find_by_capability. Ranking basis: relevance score desc (+1 per term hit, +3 for full-phrase hit); within an equal relevance tier, callable-now (Open access) + remote + healthy tools are boosted ahead — never a hard filter (a product-bound tool stays in its own money query), then productName. Paid tier is NOT a ranking input — it appears only as an annotation. Every result carries { position (1-based), rank (0..1; 1.0 = top, scales linearly down by ordinal position) } plus preflight annotations { accessModel (open|keyed|account|commercial), healthScore (0-100, nullable), readOnly, destructive, callableNow (true = usable anonymously right now), setup (one-line 'how to get access' pointer for non-Open tools) }. Response: { task, matchType (none|exactSlug|canonicalSlug|nlpFallback), resolvedCapabilities[], capabilityProducts[] (B2B SaaS products that claim the same capability — empty when matchType=none), results[] (MCP-tool-level matches), buckets { callableNow, requiresSetup } }. Each result: { position, rank, normalizedName, displayName, description, inputSchema, relevance, accessModel, healthScore, readOnly, destructive, callableNow, setup, productSlug, productName, serverQualifiedName, isRemoteCapable, tier, unverified, verifiedAt }. Empty case: { task, matchType, message, suggestedQueries[], suggestedCapabilities[] }.