Adw Bootstrap — Development skill for Claude Code
Scan this repo and write the context files every ADW stage reads first.
How to install Adw Bootstrap
Installs to ~/.claude/commands/frostyfruit-adw-kit-adw-bootstrap.md
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/FrostyFruit/adw-kit/HEAD/.claude/commands/adw-bootstrap.md -o ~/.claude/commands/frostyfruit-adw-kit-adw-bootstrap.md Restart Claude Code, or start a new session, for it to be picked up.
What Adw Bootstrap does
description: Scan this repo and write the context files every ADW stage reads first
You are setting up the AI Developer Workflow (ADW) for THIS repository. Your job is discovery and writing context files. You are not implementing features.
The stage prompts (`adw-scout` through `adw-handover`) are the same for every repo. What makes the loop fit THIS codebase is the context files you write here. Get them right and every future run starts from what this repo has already learned.
Pha
Alternatives in Development
- Setup Context — Set up your Skill Memory — a pm-context.md every skill reads so outputs come back tailored to you 1.3k ★
- Arstart — AgentRecall session opener — status board (no args), project loader ( ), or cold-start bootstrap 370 ★
- 02a Empirical Reality Check — Stage 2a: Empirical Reality Check / Institutional Context Check 164 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Graph Init
Scan this repo and write a graph-engineering profile - the stack map, skill routing table, rule paths, gates a
Reload Session Bootstrap
The first thing an agent should do each session in Reload — load its shared-memory context, then find the mess
Delegate First
Keep the main conversation clean by forking implementation work to sub-agents. Use when you want ongoing conve
Otel Init
First-run: scan the repo, detect service boundaries, and prime the context cache
Adw Handover
ADW step 7 - write the handover and teach the codebase what this run learned
Adw Review
ADW step 6 - a FRESH agent that did not write the change judges it against the criteria
Related Agents
Cook Audit
jeff audit stage (conditional: runs when the plan flags a security-relevant surface, or when the mechanical sc
Orchestration Implementer
Implements exactly one ticket end-to-end in an isolated git worktree, TDD red-green, then opens a PR to the co
Git Pull
Safely pull updates (fork/upstream aware), stash-first, resolve conflicts via memory-bank context, no auto-sta
