# How to integrate Countdown api MCP with Mastra AI ```json { "title": "How to integrate Countdown api MCP with Mastra AI", "toolkit": "Countdown api", "toolkit_slug": "countdown_api", "framework": "Mastra AI", "framework_slug": "mastra-ai", "url": "https://composio.dev/toolkits/countdown_api/framework/mastra-ai", "markdown_url": "https://composio.dev/toolkits/countdown_api/framework/mastra-ai.md", "updated_at": "2026-05-12T10:07:41.168Z" } ``` ## Introduction This guide walks you through connecting Countdown api to Mastra AI using the Composio tool router. By the end, you'll have a working Countdown api agent that can list all your ebay data collections, start processing requests for a collection, get autocomplete suggestions for 'wireless earbuds' through natural language commands. This guide will help you understand how to give your Mastra AI agent real control over a Countdown api account through Composio's Countdown api MCP server. Before we dive in, let's take a quick look at the key ideas and tools involved. ## Also integrate Countdown api with - [OpenAI Agents SDK](https://composio.dev/toolkits/countdown_api/framework/open-ai-agents-sdk) - [Claude Agent SDK](https://composio.dev/toolkits/countdown_api/framework/claude-agents-sdk) - [Claude Code](https://composio.dev/toolkits/countdown_api/framework/claude-code) - [Claude Cowork](https://composio.dev/toolkits/countdown_api/framework/claude-cowork) - [Codex](https://composio.dev/toolkits/countdown_api/framework/codex) - [OpenClaw](https://composio.dev/toolkits/countdown_api/framework/openclaw) - [Hermes](https://composio.dev/toolkits/countdown_api/framework/hermes-agent) - [CLI](https://composio.dev/toolkits/countdown_api/framework/cli) - [Google ADK](https://composio.dev/toolkits/countdown_api/framework/google-adk) - [LangChain](https://composio.dev/toolkits/countdown_api/framework/langchain) - [Vercel AI SDK](https://composio.dev/toolkits/countdown_api/framework/ai-sdk) - [LlamaIndex](https://composio.dev/toolkits/countdown_api/framework/llama-index) - [CrewAI](https://composio.dev/toolkits/countdown_api/framework/crew-ai) ## TL;DR Here's what you'll learn: - Set up your environment so Mastra, OpenAI, and Composio work together - Create a Tool Router session in Composio that exposes Countdown api tools - Connect Mastra's MCP client to the Composio generated MCP URL - Fetch Countdown api tool definitions and attach them as a toolset - Build a Mastra agent that can reason, call tools, and return structured results - Run an interactive CLI where you can chat with your Countdown api agent ## What is Mastra AI? Mastra AI is a TypeScript framework for building AI agents with tool support. It provides a clean API for creating agents that can use external services through MCP. Key features include: - MCP Client: Built-in support for Model Context Protocol servers - Toolsets: Organize tools into logical groups - Step Callbacks: Monitor and debug agent execution - OpenAI Integration: Works with OpenAI models via @ai-sdk/openai ## What is the Countdown api MCP server, and what's possible with it? The Countdown api MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Countdown api account. It provides structured and secure access to real-time eBay marketplace data, so your agent can perform actions like searching eBay products, managing collections, retrieving seller feedback, and automating product data workflows on your behalf. - eBay product search and autocomplete: Instantly fetch eBay autocomplete suggestions and help agents surface relevant product search terms and ideas in real time. - Collection management and orchestration: Create, update, list, or delete collections to batch and organize multiple eBay data requests for streamlined marketplace analysis. - Automated collection processing: Start or clear queued requests within a collection, making it easy to control and automate data gathering operations from eBay. - Destination setup and notifications: Set up or remove destinations for results and notifications, ensuring your agent can manage where and how you receive processed eBay data. - Access to rich eBay metadata: Retrieve detailed collection information, product details, customer reviews, and seller feedback to power analytics and business decisions. ## Supported Tools | Tool slug | Name | Description | |---|---|---| | `COUNTDOWN_API_CLEAR_ALL_REQUESTS` | Clear Collection Requests | Clears (removes) all pending requests from a collection. Use this to reset a collection before adding new requests, or to cancel all queued requests. The collection must be in 'idle' status. Returns the updated collection details after clearing. | | `COUNTDOWN_API_COLLECTIONS_CREATE_COLLECTION` | Create a new collection | Tool to create a new collection. Use when you need to batch and orchestrate multiple requests on a schedule. | | `COUNTDOWN_API_COLLECTIONS_GET_COLLECTION` | Get Collection | Tool to retrieve details for a single collection by ID. Use when you need collection metadata and counts for a given collection ID. | | `COUNTDOWN_API_COLLECTIONS_LIST_COLLECTIONS` | List Collections | Tool to list all collections for the authenticated account. Use when you need to retrieve paginated collections with filtering, sorting, and timing constraints. | | `COUNTDOWN_API_COLLECTIONS_START_COLLECTION` | Start Collection | Start processing a collection's queued requests on the Countdown API. Use this tool to manually trigger a collection to begin executing its requests. The collection must have at least one request added to it before it can be started. You need sufficient API credits to run all requests in the collection. Prerequisites: - Collection must exist (use List Collections or Create Collection first) - Collection must have at least one request added - Sufficient API credits available for the number of requests - Collection must not already be running | | `COUNTDOWN_API_COLLECTIONS_UPDATE_COLLECTION` | Update an existing collection | Update an existing collection's settings. Use this to modify collection properties like name, schedule, priority, notification settings, or enabled status. The collection_id must be obtained from list_collections or create_collection. Only specified fields are updated; omitted fields retain their current values. | | `COUNTDOWN_API_CORE_API_SEND_REQUEST` | eBay Autocomplete | Tool to fetch eBay autocomplete suggestions. Use when needing search-term-based suggestions from Countdown API. | | `COUNTDOWN_API_CREATE_COLLECTION_REQUEST` | Create Collection Request | Tool to create new requests within a collection for bulk eBay data retrieval. Use when you need to add search, product, reviews, or other eBay data queries to a collection that will be executed on schedule. Collections must be in 'idle' status to accept new requests. | | `COUNTDOWN_API_CREATE_DESTINATION` | Create Destination | Creates a cloud storage destination where batch result sets will be automatically uploaded. Supports Amazon S3, S3-compatible services (MinIO, DigitalOcean Spaces), Google Cloud Storage, Microsoft Azure Blob Storage, and Alibaba Cloud OSS. When enabled=true, the API validates credentials by uploading and deleting a test file. Set enabled=false to create the destination without credential validation. | | `COUNTDOWN_API_DELETE_COLLECTION` | Delete Collection | Tool to delete a collection and its configuration by ID. Use when you need to remove an existing, non-running collection. | | `COUNTDOWN_API_DELETE_DESTINATION` | Delete Destination | Tool to delete a destination by ID. Use when you need to remove an existing destination. | | `COUNTDOWN_API_DELETE_SINGLE_REQUEST` | Delete Single Request | Delete a specific request from a Countdown API collection by its ID. Use this to remove individual requests that are no longer needed. The collection must not be running when deleting requests. | | `COUNTDOWN_API_DESTINATIONS_LIST_DESTINATIONS` | List Destinations | Tool to list all destinations configured for the account. Use when you need to inspect or paginate through configured destinations. | | `COUNTDOWN_API_FIND_COLLECTION_REQUESTS` | Find Collection Requests | Tool to find requests in a collection by custom_id or search query. Use when you need to search for specific requests within a collection using either an exact custom_id match or a text search query. | | `COUNTDOWN_API_GET_ACCOUNT` | Get Account Information | Tool to retrieve account usage and current platform status. Use when needing to check plan, usage, and quota details for the authenticated user. | | `COUNTDOWN_API_LIST_ERROR_LOGS` | List Error Logs | Tool to list error logs from collection executions. Returns recent errors encountered during request processing with details about failed requests and their causes. | | `COUNTDOWN_API_REQUESTS_EXPORT_CSV` | Export Requests CSV | Export all requests from a collection as downloadable CSV files. Returns URLs to CSV files containing the request data. Use this tool when you need to bulk export or download collection request data in CSV format. The response includes paginated download links if the collection has many requests. Collections with no requests will return an empty pages array. | | `COUNTDOWN_API_REQUESTS_EXPORT_JSON` | Export Requests as JSON | Tool to download all requests in a collection as JSON. Use when you need to export the entire request history for a collection. | | `COUNTDOWN_API_REQUESTS_UPDATE_SINGLE_REQUEST` | Update Single Request | Tool to modify parameters of an existing request in a collection. Use when the collection is not running and you need to update eBay Product Data API parameters. | | `COUNTDOWN_API_RESULTS_GET_RESULT_SET` | Get Result Set | Tool to retrieve a collection run's result set payload. Use after a collection run completes to fetch metadata and download links. | | `COUNTDOWN_API_RESULTS_LIST_RESULT_SETS` | List Result Sets | Tool to list result sets produced by a collection. Use when you need to retrieve all summary status of result sets generated by a collection within the 14-day retention window. | | `COUNTDOWN_API_RESULTS_RESEND_RESULT_SET_WEBHOOK` | Resend Result Set Webhook | Resend the webhook notification for a collection's result set. Use this tool to retry webhook delivery when the original webhook POST failed or timed out. The collection must have a notification_webhook URL configured (either on the collection itself or on the account profile). Result sets are only available for 14 days after creation. Prerequisites: - Collection must exist with a valid notification_webhook URL configured - Result set must exist and not be expired (14-day retention) - Use List Result Sets to find valid result_set_id values | | `COUNTDOWN_API_STOP_ALL_COLLECTIONS` | Stop All Collections | Tool to stop all collections. Use when you need to halt any running or queued collections after reviewing operations. | | `COUNTDOWN_API_STOP_COLLECTION` | Stop Collection | Tool to stop (pause) a single collection’s processing by ID. Use when you need to halt a running or queued collection after confirming the target collection ID. | | `COUNTDOWN_API_UPDATE_DESTINATION` | Update Destination | Tool to update a destination's configuration by ID. Use after creating or retrieving a destination to modify its settings. | ## Supported Triggers None listed. ## Creating MCP Server - Stand-alone vs Composio SDK The Countdown api MCP server is an implementation of the Model Context Protocol that connects your AI agent to Countdown api. It provides structured and secure access so your agent can perform Countdown api operations on your behalf through a secure, permission-based interface. With Composio's managed implementation, you don't have to create your own developer app. For production, if you're building an end product, we recommend using your own credentials. The managed server helps you prototype fast and go from 0-1 faster. ## Step-by-step Guide ### 1. Prerequisites Before starting, make sure you have: - Node.js 18 or higher - A Composio account with an active API key - An OpenAI API key - Basic familiarity with TypeScript ### 1. Getting API Keys for OpenAI and Composio OpenAI API Key - Go to the [OpenAI dashboard](https://platform.openai.com/settings/organization/api-keys) and create an API key. - You need credits or a connected billing setup to use the models. - Store the key somewhere safe. Composio API Key - Log in to the [Composio dashboard](https://dashboard.composio.dev?utm_source=toolkits&utm_medium=framework_docs). - Go to Settings and copy your API key. - This key lets your Mastra agent talk to Composio and reach Countdown api through MCP. ### 2. Install dependencies Install the required packages. What's happening: - @composio/core is the Composio SDK for creating MCP sessions - @mastra/core provides the Agent class - @mastra/mcp is Mastra's MCP client - @ai-sdk/openai is the model wrapper for OpenAI - dotenv loads environment variables from .env ```bash npm install @composio/core @mastra/core @mastra/mcp @ai-sdk/openai dotenv ``` ### 3. Set up environment variables Create a .env file in your project root. What's happening: - COMPOSIO_API_KEY authenticates your requests to Composio - COMPOSIO_USER_ID tells Composio which user this session belongs to - OPENAI_API_KEY lets the Mastra agent call OpenAI models ```bash COMPOSIO_API_KEY=your_composio_api_key_here COMPOSIO_USER_ID=your_user_id_here OPENAI_API_KEY=your_openai_api_key_here ``` ### 4. Import libraries and validate environment What's happening: - dotenv/config auto loads your .env so process.env.* is available - openai gives you a Mastra compatible model wrapper - Agent is the Mastra agent that will call tools and produce answers - MCPClient connects Mastra to your Composio MCP server - Composio is used to create a Tool Router session ```typescript import "dotenv/config"; import { openai } from "@ai-sdk/openai"; import { Agent } from "@mastra/core/agent"; import { MCPClient } from "@mastra/mcp"; import { Composio } from "@composio/core"; import * as readline from "readline"; import type { AiMessageType } from "@mastra/core/agent"; const openaiAPIKey = process.env.OPENAI_API_KEY; const composioAPIKey = process.env.COMPOSIO_API_KEY; const composioUserID = process.env.COMPOSIO_USER_ID; if (!openaiAPIKey) throw new Error("OPENAI_API_KEY is not set"); if (!composioAPIKey) throw new Error("COMPOSIO_API_KEY is not set"); if (!composioUserID) throw new Error("COMPOSIO_USER_ID is not set"); const composio = new Composio({ apiKey: composioAPIKey as string, }); ``` ### 5. Create a Tool Router session for Countdown api What's happening: - create spins up a short-lived MCP HTTP endpoint for this user - The toolkits array contains "countdown_api" for Countdown api access - session.mcp.url is the MCP URL that Mastra's MCPClient will connect to ```typescript async function main() { const session = await composio.create( composioUserID as string, { toolkits: ["countdown_api"], }, ); const composioMCPUrl = session.mcp.url; console.log("Countdown api MCP URL:", composioMCPUrl); ``` ### 6. Configure Mastra MCP client and fetch tools What's happening: - MCPClient takes an id for this client and a list of MCP servers - The headers property includes the x-api-key for authentication - getTools fetches the tool definitions exposed by the Countdown api toolkit ```typescript const mcpClient = new MCPClient({ id: composioUserID as string, servers: { nasdaq: { url: new URL(composioMCPUrl), requestInit: { headers: session.mcp.headers, }, }, }, timeout: 30_000, }); console.log("Fetching MCP tools from Composio..."); const composioTools = await mcpClient.getTools(); console.log("Number of tools:", Object.keys(composioTools).length); ``` ### 7. Create the Mastra agent What's happening: - Agent is the core Mastra agent - name is just an identifier for logging and debugging - instructions guide the agent to use tools instead of only answering in natural language - model uses openai("gpt-5") to configure the underlying LLM ```typescript const agent = new Agent({ name: "countdown_api-mastra-agent", instructions: "You are an AI agent with Countdown api tools via Composio.", model: "openai/gpt-5", }); ``` ### 8. Set up interactive chat interface What's happening: - messages keeps the full conversation history in Mastra's expected format - agent.generate runs the agent with conversation history and Countdown api toolsets - maxSteps limits how many tool calls the agent can take in a single run - onStepFinish is a hook that prints intermediate steps for debugging ```typescript let messages: AiMessageType[] = []; console.log("Chat started! Type 'exit' or 'quit' to end.\n"); const rl = readline.createInterface({ input: process.stdin, output: process.stdout, prompt: "> ", }); rl.prompt(); rl.on("line", async (userInput: string) => { const trimmedInput = userInput.trim(); if (["exit", "quit", "bye"].includes(trimmedInput.toLowerCase())) { console.log("\nGoodbye!"); rl.close(); process.exit(0); } if (!trimmedInput) { rl.prompt(); return; } messages.push({ id: crypto.randomUUID(), role: "user", content: trimmedInput, }); console.log("\nAgent is thinking...\n"); try { const response = await agent.generate(messages, { toolsets: { countdown_api: composioTools, }, maxSteps: 8, }); const { text } = response; if (text && text.trim().length > 0) { console.log(`Agent: ${text}\n`); messages.push({ id: crypto.randomUUID(), role: "assistant", content: text, }); } } catch (error) { console.error("\nError:", error); } rl.prompt(); }); rl.on("close", async () => { console.log("\nSession ended."); await mcpClient.disconnect(); process.exit(0); }); } main().catch((err) => { console.error("Fatal error:", err); process.exit(1); }); ``` ## Complete Code ```typescript import "dotenv/config"; import { openai } from "@ai-sdk/openai"; import { Agent } from "@mastra/core/agent"; import { MCPClient } from "@mastra/mcp"; import { Composio } from "@composio/core"; import * as readline from "readline"; import type { AiMessageType } from "@mastra/core/agent"; const openaiAPIKey = process.env.OPENAI_API_KEY; const composioAPIKey = process.env.COMPOSIO_API_KEY; const composioUserID = process.env.COMPOSIO_USER_ID; if (!openaiAPIKey) throw new Error("OPENAI_API_KEY is not set"); if (!composioAPIKey) throw new Error("COMPOSIO_API_KEY is not set"); if (!composioUserID) throw new Error("COMPOSIO_USER_ID is not set"); const composio = new Composio({ apiKey: composioAPIKey as string }); async function main() { const session = await composio.create(composioUserID as string, { toolkits: ["countdown_api"], }); const composioMCPUrl = session.mcp.url; const mcpClient = new MCPClient({ id: composioUserID as string, servers: { countdown_api: { url: new URL(composioMCPUrl), requestInit: { headers: session.mcp.headers, }, }, }, timeout: 30_000, }); const composioTools = await mcpClient.getTools(); const agent = new Agent({ name: "countdown_api-mastra-agent", instructions: "You are an AI agent with Countdown api tools via Composio.", model: "openai/gpt-5", }); let messages: AiMessageType[] = []; const rl = readline.createInterface({ input: process.stdin, output: process.stdout, prompt: "> ", }); rl.prompt(); rl.on("line", async (input: string) => { const trimmed = input.trim(); if (["exit", "quit"].includes(trimmed.toLowerCase())) { rl.close(); return; } messages.push({ id: crypto.randomUUID(), role: "user", content: trimmed }); const { text } = await agent.generate(messages, { toolsets: { countdown_api: composioTools }, maxSteps: 8, }); if (text) { console.log(`Agent: ${text}\n`); messages.push({ id: crypto.randomUUID(), role: "assistant", content: text }); } rl.prompt(); }); rl.on("close", async () => { await mcpClient.disconnect(); process.exit(0); }); } main(); ``` ## Conclusion You've built a Mastra AI agent that can interact with Countdown api through Composio's Tool Router. You can extend this further by: - Adding other toolkits like Gmail, Slack, or GitHub - Building a web-based chat interface around this agent - Using multiple MCP endpoints to enable cross-app workflows ## How to build Countdown api MCP Agent with another framework - [OpenAI Agents SDK](https://composio.dev/toolkits/countdown_api/framework/open-ai-agents-sdk) - [Claude Agent SDK](https://composio.dev/toolkits/countdown_api/framework/claude-agents-sdk) - [Claude Code](https://composio.dev/toolkits/countdown_api/framework/claude-code) - [Claude Cowork](https://composio.dev/toolkits/countdown_api/framework/claude-cowork) - [Codex](https://composio.dev/toolkits/countdown_api/framework/codex) - [OpenClaw](https://composio.dev/toolkits/countdown_api/framework/openclaw) - [Hermes](https://composio.dev/toolkits/countdown_api/framework/hermes-agent) - [CLI](https://composio.dev/toolkits/countdown_api/framework/cli) - [Google ADK](https://composio.dev/toolkits/countdown_api/framework/google-adk) - [LangChain](https://composio.dev/toolkits/countdown_api/framework/langchain) - [Vercel AI SDK](https://composio.dev/toolkits/countdown_api/framework/ai-sdk) - [LlamaIndex](https://composio.dev/toolkits/countdown_api/framework/llama-index) - [CrewAI](https://composio.dev/toolkits/countdown_api/framework/crew-ai) ## Related Toolkits - [Addresszen](https://composio.dev/toolkits/addresszen) - Addresszen is a real-time address autocomplete and verification service. It helps capture accurate, deliverable addresses with instant suggestions and validation. - [Asin data api](https://composio.dev/toolkits/asin_data_api) - Asin data api gives you detailed, real-time product data from Amazon, including price, rank, and reviews. Perfect for e-commerce pros and data-driven marketers who need instant marketplace insights. - [Baselinker](https://composio.dev/toolkits/baselinker) - BaseLinker is an all-in-one e-commerce management platform connecting stores, marketplaces, carriers, and more. It streamlines order processing, inventory control, and automates your sales operations. - [Bestbuy](https://composio.dev/toolkits/bestbuy) - Best Buy is a leading retailer offering APIs for product, store, and recommendation data. Instantly access up-to-date retail insights for smarter shopping and decision-making. - [Btcpay server](https://composio.dev/toolkits/btcpay_server) - BTCPay Server is a free, open-source, self-hosted Bitcoin payment processor. It lets merchants accept Bitcoin payments directly, cutting out middlemen and boosting privacy. - [Cdr platform](https://composio.dev/toolkits/cdr_platform) - Cdr platform is an API for purchasing carbon dioxide removal services. It enables businesses to offset emissions by accessing verified carbon removal projects. - [Cloudcart](https://composio.dev/toolkits/cloudcart) - CloudCart is an e-commerce platform for building and managing online stores. It helps businesses streamline product listings, orders, and customer engagement. - [Dpd2](https://composio.dev/toolkits/dpd2) - Dpd2 is a robust email management platform for handling, sorting, and automating email workflows. Streamline your communications and boost productivity with advanced sorting, labeling, and response tools. - [Finerworks](https://composio.dev/toolkits/finerworks) - FinerWorks is an online platform for fine art and photo printing services. Artists and photographers use it to order custom prints and manage print inventory efficiently. - [Fingertip](https://composio.dev/toolkits/fingertip) - Fingertip is a business management platform for selling, booking, and customer engagement—all from a single link. It helps businesses streamline operations and connect with customers across social channels. - [Fraudlabs pro](https://composio.dev/toolkits/fraudlabs_pro) - FraudLabs Pro is an online payment fraud detection service for e-commerce and merchants. It helps minimize chargebacks and revenue loss by detecting and preventing fraudulent transactions. - [Gift up](https://composio.dev/toolkits/gift_up) - Gift Up! is a digital platform for selling, managing, and redeeming gift cards online. It streamlines promotions and gift card transactions for businesses and their customers. - [Goody](https://composio.dev/toolkits/goody) - Goody is a gifting platform that lets users send gifts and physical products without handling logistics. It streamlines gifting by managing delivery, fulfillment, and recipient experience. - [Gumroad](https://composio.dev/toolkits/gumroad) - Gumroad is a platform for selling digital products, physical goods, and memberships with a simple checkout and marketing tools. It streamlines creator payouts and helps you grow your audience effortlessly. - [Instacart](https://composio.dev/toolkits/instacart) - Instacart is an online grocery delivery and pickup service platform. It lets you discover local retailers and create shoppable lists and recipes with ease. - [Junglescout](https://composio.dev/toolkits/junglescout) - Junglescout is an Amazon product research and analytics platform for sellers. It delivers sales estimates, competitive insights, and optimization tools to boost your Amazon business. - [Ko fi](https://composio.dev/toolkits/ko_fi) - Ko-fi is a platform that lets creators receive donations, memberships, and sales from fans. It helps creators monetize their work and grow their audience with minimal friction. - [Lemon squeezy](https://composio.dev/toolkits/lemon_squeezy) - Lemon Squeezy is a payments and subscription platform built for software companies. It makes managing payments, taxes, and customer subscriptions effortless. - [Loyverse](https://composio.dev/toolkits/loyverse) - Loyverse is a point-of-sale (POS) platform for small businesses, offering tools for sales, inventory, and customer loyalty. It helps streamline retail operations and boost customer engagement. - [Memberstack](https://composio.dev/toolkits/memberstack) - Memberstack lets you add user authentication, payments, and member management to your website—no backend code required. Easily manage your site's members and subscriptions from a single platform. ## Frequently Asked Questions ### What are the differences in Tool Router MCP and Countdown api MCP? With a standalone Countdown api MCP server, the agents and LLMs can only access a fixed set of Countdown api tools tied to that server. However, with the Composio Tool Router, agents can dynamically load tools from Countdown api and many other apps based on the task at hand, all through a single MCP endpoint. ### Can I use Tool Router MCP with Mastra AI? Yes, you can. Mastra AI fully supports MCP integration. You get structured tool calling, message history handling, and model orchestration while Tool Router takes care of discovering and serving the right Countdown api tools. ### Can I manage the permissions and scopes for Countdown api while using Tool Router? Yes, absolutely. You can configure which Countdown api scopes and actions are allowed when connecting your account to Composio. You can also bring your own OAuth credentials or API configuration so you keep full control over what the agent can do. ### How safe is my data with Composio Tool Router? All sensitive data such as tokens, keys, and configuration is fully encrypted at rest and in transit. Composio is SOC 2 Type 2 compliant and follows strict security practices so your Countdown api data and credentials are handled as safely as possible. --- [See all toolkits](https://composio.dev/toolkits) · [Composio docs](https://docs.composio.dev/llms.txt)