Skip to content

Latest commit

 

History

History
51 lines (32 loc) · 1.83 KB

AGENTS.md

File metadata and controls

51 lines (32 loc) · 1.83 KB

AGENTS.md

Instructions for AI coding agents working with this codebase.

Source Code Reference

Source code for dependencies is available in opensrc/ for deeper understanding of implementation details.

See opensrc/sources.json for the list of available packages and their versions.

Use this source code when you need to understand how a package works internally, not just its types/interface.

Fetching Additional Source Code

To fetch source code for a package or repository you need to understand, run:

npx opensrc <package>           # npm package (e.g., npx opensrc zod)
npx opensrc pypi:<package>      # Python package (e.g., npx opensrc pypi:requests)
npx opensrc crates:<package>    # Rust crate (e.g., npx opensrc crates:serde)
npx opensrc <owner>/<repo>      # GitHub repo (e.g., npx opensrc vercel/ai)

llms-full reference

When working on tasks about a library/framework/runtime/platform, first consult llms-furl/, which contains llms-full.txt split into a tree of leaves — small, searchable files for quick lookup.

Workflow:

  1. Check domains in llms-furl/AGENTS.md.
  2. Search within the relevant domain (e.g. rg -n "keyword" llms-furl/bun.sh).
  3. If needed, navigate with index.json using jq.
  4. If no relevant info is found, state that and then move on to other sources.

Project notes workflow (.workspace-fs/)

When you make changes that affect how the next person/agent should work (env vars, setup, API shape, state machines, key decisions), also update the relevant files under .workspace-fs/ so it remains the single source of truth:

  • .workspace-fs/state.md (current state + next tasks)
  • .workspace-fs/decisions.md (decisions + rationale)
  • .workspace-fs/worklog-YYYY-MM-DD.md (what changed today)