AgentDrives

Choose a payments, email, sign-in, scheduling or data provider for a task: send the task in a few plain words ('send an invoice') with filters if they matter, and get the SaaS, APIs, MCP servers and CLIs ranked by how far our agent got. Use search_tools when you have a name, domain or kind, and get_tool for one product. Read-only, no key needed; 5 results by default (up to 25), each with its agent level (0-5), the closest task and what happened, and next steps: sign-up, docs and MCP links.

Search the directory of driven SaaS, API, MCP server and CLI products by a product's name, domain or kind: send a short query ('stripe.com', 'postgres', 'invoicing') with filters for category, agent level, access or payment. Use find_tools_for_task instead for a task in plain words, and get_tool for one product in full. Read-only, no key needed; up to 100 results a page sorted by agent level, with total and next_offset for the rest.

Get one product's result, when the question is whether an AI agent can use it: send its slug from search_tools, or its name or domain, and get its test-drive result (agent level 0-5, where our agent stopped and why, what it can and can't do, how it gets in and pays). Use get_agent_surfaces instead when you only need its API, MCP, CLI and SDK. Read-only, no key needed; the headline (2.5 to 5 KB) unless brief is false (13 to 35 KB).

List what an agent can use at one product: its API, MCP server, CLI, SDK, docs and payment, each tested, documented, none_found or unknown, with the product's own URL for it and the date we saw it. Use it to wire up a product you already picked; use get_tool for the whole drive result. Read-only, no key needed; about 1 KB.

Compare 2 or 3 products side by side for one task, from our test drives: send their slugs (from search_tools), names or domains, and the task in a few plain words. For each: driven or only listed, agent level (0-5), how far our agent got at each step and where it stopped, what it did on the task, how to connect (MCP, API, CLI), the date and its page; a name we don't cover comes back in not_covered. Read-only, no key needed.

List every category in the directory with its id, name, number of products and best agent level (0-5). Use it for a valid category id before find_tools_for_task, search_tools or get_category. Read-only, no key needed; takes no arguments.

Takes no parameters.

List what changed in the directory lately: products whose agent level moved between two drives, with the date, the cause and the level now, newest first. Send a slug to get every dated change at one product you picked (level, surfaces that appeared or stopped answering, price and plan, each with its source) and see whether it still works the same; use get_tool for its full drive result. Read-only, no key needed; narrow it with days, category or slug.

Get one category's leaders: its name, how many products AgentDrives drove in it, the best agent level in it (0-5) and its top 10 products by level, each with its drive date and page. Use it to shortlist in a category you know; use search_tools with the same category for the rest. Read-only, no key needed.

Answers "which tool works best for X?": one category's driven products ranked by agent level (0-5), each with where our agent stopped, and how many we drove in it; under 5, not enough drives yet. Then get_tool for one product. Read-only, no key needed.