# Utility Llms API (https://www.socialcrawl.dev/platforms/utility/llms) > The SocialCrawl context corpus for AI agents, served through the API: the same content as llms.txt for the whole API or a single platform, as markdown or structured JSON. An agent with an API key can bootstrap itself with one call instead of scraping documentation pages. Free to call. TL;DR: `GET /v1/utility/llms` costs Free (0 credits) and returns SocialCrawl's unified JSON schema. Single x-api-key auth, 100 free credits on signup. ## Parameters | Parameter | Required | Description | | --- | --- | --- | | `platform` | no | Scope the context to one platform slug | | `format` | no | markdown (default) returns the llms corpus text; json returns a structured context object (markdown \| json) | ## Code example ```bash curl "https://www.socialcrawl.dev/v1/utility/llms" \ -H "x-api-key: sc_YOUR_API_KEY" ``` ## FAQ ### What does GET /v1/utility/llms return? The SocialCrawl context corpus for AI agents, served through the API: the same content as the public llms.txt file, for the whole API or scoped to a single platform, as markdown by default or as a structured JSON context object with format=json. ### Why fetch llms.txt through the API instead of the file? An agent that already holds an API key bootstraps itself with one authenticated call and gets the corpus in the same envelope as every other response. The json format also gives it a structured object instead of raw text, which is easier to slot into a context window. ### Can I scope the agent context to one platform? Yes. Pass platform (for example platform=tiktok) and the corpus covers just that platform's endpoints and rules, which keeps the payload small when an agent only works with one source. ### Is the llms endpoint free to call? Yes, it costs 0 credits, like the endpoints catalog, the endpoint usage guide, and the quickstart. Loading context never spends credits, so agents can refresh their knowledge of the API whenever they need to. ### How do AI agents self-onboard with SocialCrawl? With the four free utility endpoints: quickstart for auth, billing, and error rules, llms for the context corpus, endpoints to discover the catalog, and endpoint for exact per-endpoint usage. An agent with an API key runs that loop on its own, with no documentation scraping and no credits spent. See the full Utility API: https://www.socialcrawl.dev/platforms/utility ## Pricing - Standard endpoints: 1 credit per call - Advanced endpoints: 5 credits per call - Premium endpoints: 10 credits per call - 100 free credits on signup, no credit card required. Cached responses cost 0 credits. Credit packs never expire. - Full pricing: https://www.socialcrawl.dev/pricing ## Explore with AI Questions this API answers, phrased for an AI assistant: - How can an AI agent load SocialCrawl's llms.txt through the API? - Which API serves LLM agent context as structured JSON?