# Aa Resume

> Use when the user says resume, continue, or pick up where we left off, starts a new session on an existing Agent-Agile project, or is recovering after a crash or context reset, for example "/aa-resume", "let's continue", "where were we".

- **Type:** Skill
- **Install:** `agentstack add skill-peeyushmeher-agent-agile-aa-resume`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [Peeyushmeher](https://agentstack.voostack.com/s/peeyushmeher)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [Peeyushmeher](https://github.com/Peeyushmeher)
- **Source:** https://github.com/Peeyushmeher/agent-agile/tree/master/skills/aa-resume

## Install

```sh
agentstack add skill-peeyushmeher-agent-agile-aa-resume
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# aa-resume

Resolve the Agent-Agile playbook root: use the first of these that exists — (1) `${CLAUDE_PLUGIN_ROOT}/playbooks`, (2) `./.claude/agent-agile/playbooks`, (3) `./.agents/agent-agile/playbooks`, (4) `~/.claude/agent-agile/playbooks`, (5) `~/.agents/agent-agile/playbooks`, (6) `./playbooks`.

Read `playbooks/execution.md`'s "Resume protocol" section from that root, and follow it exactly:

1. If `.planning/STATE.md` does not exist, no project has been started here yet. Say so plainly, and point the user at the three front doors instead of guessing at a state that isn't there: a vague spark → `/aa-grill`, a clear idea → `/aa-new-project`, an existing plan or spec → `/aa-import`.
2. Otherwise, read `.planning/STATE.md` first and only. Don't re-derive status by re-reading the whole epic history — `STATE.md` is the baton, kept small on purpose so this step stays cheap every time.
3. Follow its pointers — at most three files — and read exactly those, nothing wider.
4. Continue the work from exactly where `STATE.md` says the run is: the phase it names (grill, intake, panel, prereqs, an epic's wave, review, or done), the "Now" sentence, and the "Next" sentence.
5. If `STATE.md` names a blocker, resolve or surface that blocker before doing anything else — never work around it silently.

Close the session, without exception, with one `STATE.md` update: what's in progress, what's next, the pointers the next session needs, and any blocker. A session that stops without this update leaves the next session to reconstruct state from scratch — the entire point of the baton is that this never has to happen.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [Peeyushmeher](https://github.com/Peeyushmeher)
- **Source:** [Peeyushmeher/agent-agile](https://github.com/Peeyushmeher/agent-agile)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-peeyushmeher-agent-agile-aa-resume
- Seller: https://agentstack.voostack.com/s/peeyushmeher
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.