How do I install the Aidd Async Dev:03:review skill?

Run npx skillmds@latest add ai-driven-dev/aidd-async-dev-03-review in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Aidd Async Dev:03:review safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Aidd Async Dev:03:review?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Aidd Async Dev:03:review free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Aidd Async Dev:03:review?

ai-driven-dev (@ai-driven-dev) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Aidd Async Dev:03:review by ai-driven-dev

Name: Aidd Async Dev:03:review
Author: ai-driven-dev

Aidd Async Dev:03:review

Handles the post-PR review-fix loop for runs created by this plugin's run skill. Collects review comments, decides whether to keep iterating, delegates fixes to whichever SDLC capability is loaded at runtime, and finalizes when stop conditions trigger. Use when the user (or a webhook / cron) says "handle review comments", "iterate on PR <n>", "address review feedback automatically", or invokes this skill on a specific PR. Do NOT use for the initial implementation or for setup; other skills in this plugin cover those.

ai-driven-dev a036886 7 files · 10.1 KB Updated Sep 17, 2026

File contents

Review

Closes the loop after a PR is opened by this plugin's run skill. Detects when to keep auto-fixing and when to hand off to a human.

Available actions

#	Action	Role	Input
01	`collect-comments`	Fetch PR review comments since the last iteration	PR number
02	`detect-stop`	Decide stop vs continue (max-N, label, human)	comments + state
03	`fix-iteration`	Delegate fixes to the discovered SDLC capability	comments
04	`finalize`	Update audit record and stop the loop cleanly	iteration log

Default flow

Loop: 01 -> 02. If 02 says continue, run 03 then jump back to 01. If 02 says stop, run 04 once and exit.

Transversal rules

Read .claude/aidd-async-dev.json for max_iterations and label names.
Track iteration count in the audit record produced by this plugin's run skill. Resume the same run_id, do not create a new one.
Discover the SDLC capability the same way the run skill does: search loaded skills by description. Never hardcode a skill name.
Never auto-merge. The loop only adds commits to the existing PR branch.
See references/stop-conditions.md for the precise stop logic.

References

references/stop-conditions.md -- ordered stop conditions and human detection heuristics

External data

aidd_docs/async-runs/ -- run history (this skill appends iteration entries to existing records)
.claude/aidd-async-dev.json -- runtime config produced by this plugin's setup skill

---name: aidd-async-dev-03-reviewdescription: Handles the post-PR review-fix loop for runs created by this plugin's run skill. Collects review comments, decides whether to keep iterating, delegates fixes to whichever SDLC capability is loaded at runtime, and finalizes when stop conditions trigger. Use when the user (or a webhook / cron) says "handle review comments", "iterate on PR <n>", "address review feedback automatically", or invokes this skill on a specific PR. Do NOT use for the initial implementation or for setup; other skills in this plugin cover those.---# ReviewCloses the loop after a PR is opened by this plugin's run skill. Detects when to keep auto-fixing and when to hand off to a human.## Available actions| # | Action | Role | Input || --- | ------------------- | -------------------------------------------------------------- | ------------ || 01 | `collect-comments` | Fetch PR review comments since the last iteration | PR number || 02 | `detect-stop` | Decide stop vs continue (max-N, label, human) | comments + state || 03 | `fix-iteration` | Delegate fixes to the discovered SDLC capability | comments || 04 | `finalize` | Update audit record and stop the loop cleanly | iteration log |## Default flowLoop: `01 -> 02`. If `02` says continue, run `03` then jump back to `01`. If `02` says stop, run `04` once and exit.## Transversal rules- Read `.claude/aidd-async-dev.json` for `max_iterations` and label names.- Track iteration count in the audit record produced by this plugin's run skill. Resume the same `run_id`, do not create a new one.- Discover the SDLC capability the same way the run skill does: search loaded skills by description. Never hardcode a skill name.- Never auto-merge. The loop only adds commits to the existing PR branch.- See `references/stop-conditions.md` for the precise stop logic.## References- `references/stop-conditions.md` -- ordered stop conditions and human detection heuristics## External data- `aidd_docs/async-runs/` -- run history (this skill appends iteration entries to existing records)- `.claude/aidd-async-dev.json` -- runtime config produced by this plugin's setup skill

Frequently asked questions