> ## Documentation Index > Fetch the complete documentation index at: https://tryklint.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. > ## Agent Instructions > Fetch /llms.txt for the complete page index before exploring further. Every docs page is available as markdown by appending .md to its URL. Coding agents should start at /for-agents/coding-agents.md. # Overview > Procedures your agent executes against the Klint API: paste a cookbook's URL and your context, and your agent runs the loop and returns the artifact. A cookbook is a procedure your agent executes. Each page carries the whole job: the inputs to collect, the exact API calls, the synthesis rules, and the artifact to hand back. You paste one prompt; your agent does the rest. ## Run any cookbook The pattern is the same for all of them: ```text theme={null} Fetch https://tryklint.ai/cookbooks/[name].md and execute the cookbook it contains with my context: - [the inputs that cookbook asks for] ``` Every cookbook page declares its own inputs, verifies its own output, and states its boundaries. Agents fetch the `.md` and go. ## The cookbooks Sweep a niche's recent posts, separate repeated hook patterns from one-off outliers, and return a testable brief with receipts. Discover the creators actually publishing in a niche, qualified, with every name carrying its evidence. Decompose one post into its working parts: hook, messages, offer, transcript, and performance in context. Check a concept against what's already published: who ran it, how executions differ, what's unclaimed. Extract the text layers of any post: transcript, on-screen text, and visual descriptions, cleanly assembled. Build the research set for a market: qualified accounts by size band, each with the posts that prove it belongs.