# Magicbornstudios Get Anything Done Wire Agents Md Context Bootstrap

> wire-agents-md-context-bootstrap

- Skill: `tomevault-io/magicbornstudios-get-anything-done-wire-agents-md-context-bo` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add tomevault-io/magicbornstudios-get-anything-done-wire-agents-md-context-bo`
- Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/magicbornstudios-get-anything-done-wire-agents-md-context-bo/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: tomevault-io (https://skillmd.com/u/tomevault-io)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/tomevault-io/magicbornstudios-get-anything-done-wire-agents-md-context-bo

---


# wire-agents-md-context-bootstrap

The entry contract (AGENTS.md / CLAUDE.md beside a `.planning/` dir) is
the first file every cold agent reads. Phase 11 found that drift between
that file and the live `gad` CLI surface costs ~5-10% of every session
on archaeology before the first real tool call. This skill is the audit
+ patch loop that keeps the entry contract honest.

The skill targets multi-root monorepos where every root needs a
projectid, a canonical CLI surface, and (when multi-agent) a lane map.
For single-root projects, the steps still apply — the lane map collapses
to one paragraph.

**Workflow:** [./workflow.md](./workflow.md)

## Provenance

- Source candidate: `.planning/candidates/phase-11-agents-md-and-context-system/CANDIDATE.md`
- Drafted: 2026-04-28 by `create-proto-skill`
- See [PROVENANCE.md](./PROVENANCE.md)

---
> Source: [MagicbornStudios/get-anything-done](https://github.com/MagicbornStudios/get-anything-done) — distributed by [TomeVault](https://tomevault.io).
<!-- tomevault:4.0:skill_md:2026-05-23 -->

