The FIRST thing you do, before anything else in this file — before the Awareness Context: ask **web console** (browser page, suggested) or **terminal**? Console → run `node "<D>/console.mjs" open --workflow init-update` before reading (<D>: references/web-console/ beside this SKILL.md — ~/.agents/skills/plan2code-init-update/references/web-console/ globally — or the dir in ~/.plan2code/console/console-dir), then read <D>/console.md. Every question in the Q&A flow and the restructure offer go through it. Switchable anytime. If the argument already says which — `--web` or `Use the web console for this session.` — take it and do not ask; drop the flag. Launched by the dashboard? Its session is already open — resume it (console.md → Launches), then Awareness Context. On the console, a `__stop` action is the person ending the session (console.md → Stop requests); at every session end post `finish` BEFORE `stop`; unless it is a pause, the finish carries `"dashboard": true` (the Back to the dashboard button) and you keep waiting for the press (console.md → Finishing).
Before making any updates, orient yourself to current project state: read `AGENTS.md` and `README.md`, list `src/` directory, run `ls specs/` (not Glob — gitignored) to read any active spec overview, and locate the human-docs tree (`docs/` or equivalent — don't assume).
**If missing:** "No AGENTS.md found. Create one from scratch? I can analyze the codebase and generate an initial file." Stop and wait. If yes, use `plan2code-init.md` workflow.
Structure: Progressive discovery (index + N section files)
Section files:
- .agents-docs/AGENTS-architecture.md
- .agents-docs/AGENTS-development.md
[etc.]
```
**If `.agents-docs/` does not exist:** Note: `Structure: Single-file (no .agents-docs/ directory)`
**IMPORTANT — you MUST offer restructuring.** Stop and ask:
> "Your AGENTS.md uses a single-file format. Want to restructure for progressive discovery? This splits detailed sections into `.agents-docs/` files and converts AGENTS.md to a lightweight index with summaries and links."
Wait for user response before continuing. If user accepts, restructure existing content: create `.agents-docs/` directory with section files, convert AGENTS.md to index with summaries and links. Always-inline sections (Project Overview, Git Commit Messages, How to Use This File) stay in AGENTS.md. If user declines, continue with single-file editing.
Keep every doc surface accurate and in sync. **Deep-audit surfaces this session touched; staleness-scan the rest.** Verify against actual code — never assume or fabricate.
| `README.md` + human docs | Persistent | Human — what/why, scannable |
| Active `specs/<feature>/` | Transient | Session knowledge — status, decisions, next steps. No bloat |
**Adaptive:**`AGENTS.md`/`.agents-docs/`/`specs/` are plan2code conventions — expected, but verify. `README.md` is standard. Human-docs tree varies (`docs/` or other) — detect, don't assume.
**Fix everywhere:** stale paths/commands, missing/outdated info, mistakes, redundancy.
**Never duplicate across tiers:** route each fact to its surfaces in that surface's voice — different phrasings of one truth, never copied text.
Check for other AI agent config files (`CLAUDE.md`, `GEMINI.md`, `.cursorrules`, `.github/copilot-instructions.md`, `.cursor/rules/*.md`, `.windsurf/rules/*.md`) and offer to replace them with AGENTS.md references. **No files found:** skip silently, end workflow.
> Fallback: for each detected file, offer Yes/Select/No to replace it with a pointer to AGENTS.md (warn when a file >10 lines has custom content that would be replaced). `CLAUDE.md` gets a special template opening with a `CRITICAL — MANDATORY FIRST STEP` directive to always read `AGENTS.md` (Claude Code auto-loads it); all other files get a "See AGENTS.md for complete project documentation including:" pointer with the same bullet list, using the correct relative path (`./`, `../`, or `../../` by location). For directory configs (`.cursor/rules/`, `.windsurf/rules/`), delete existing `.md` files and create a single `reference.md`.