agentleFS
AgiFlowAgiFlow/ai-plugin/skills/review-work/SKILL.md

review-work

Skill4 starsChanged 33 days agoMITView on GitHub

Quality gate: verify each acceptance criterion of a completed task/work unit, run quality checks, and create follow-up tasks for gaps. Use before merging or to audit delivered work. Invoked as /agiflow:review-work <work-unit-or-task>. Uses get_work_unit, get_task, update_task, create_task, create_task_comment.

8 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

18 rules. 0 are common, 9 are distinctive, and 9 are very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 9

    Distinctive

    This author's own choices. Worth a look.

  • 9

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Verify that completed work actually meets its acceptance criteria, catches quality issues, and is ready to ship.

    line 30
  2. Distinctive:

    Every acceptance criterion gets a pass/fail verdict with evidence.

    line 35
  3. Very specific:

    If a work unit or task slug/id is provided, load it with get_work_unit / get_task.

    line 39
  4. Very specific:

    Follow the shared AgiFlow project-management guidelines in references/agiflow-agents.md — agent assignment, the task status workflow and …

    line 45
  5. Very specific:

    Use get_work_unit MCP tool to load the work unit and its tasks.

    line 56
  6. Distinctive:

    Check devInfo for implementation evidence:

    line 77
  7. Distinctive:

    Do the changes align with the task/work unit description?

    line 95
  8. Distinctive:

    Do devInfo notes match the actual changes?

    line 97
  9. Very specific:

    Use create_task_comment to add your review findings:

    line 109
  10. Very specific:

    If work unit: use update_work_unit to confirm status "completed"

    line 138
  11. Very specific:

    If task: use update_task to confirm status "Done"

    line 139
  12. Very specific:

    Create follow-up tasks for noted improvements using batch_create_tasks

    line 145
  13. Distinctive:

    Document notes in task comment

    line 146
  14. Very specific:

    If work unit: use update_work_unit to set status back to "in_progress"

    line 150
  15. Very specific:

    If task: use update_task to set status back to "In Progress"

    line 151
  16. Distinctive:

    Document what needs to change in task comment

    line 153
  17. Distinctive:

    After reviewing all tasks in the work unit, produce a rollup:

    line 158
  18. Distinctive:

    Not creating follow-up tasks for issues found

    line 182

“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Invoked as /agiflow:review-work. In hosts without slash-prompts, this skill is triggered by matching intent and drives AgiFlow via its MCP tools.

Usage:

  • /agiflow:review-work <work-unit-slug-or-id> - Review a specific work unit
  • /agiflow:review-work <task-slug-or-id> - Review a specific task
  • /agiflow:review-work - List completed items for review

Examples:

  • /agiflow:review-work DXX-WU-1 (review work unit by slug)
  • /agiflow:review-work DXX-3 (review task by slug)
  • /agiflow:review-work (interactive selection)

Purpose Verify that completed work actually meets its acceptance criteria, catches quality issues, and is ready to ship. AI code has 1.7x more defects than human-written code — review is the last line of defense before shipping.

Guardrails

  • Review what exists — do not implement fixes during review (create follow-up tasks instead).
  • Every acceptance criterion gets a pass/fail verdict with evidence.
  • Flag issues by severity: blocker (must fix), warning (should fix), note (nice to fix).
  • Be honest — a "pass" with gaps is worse than a clear "needs rework".

If a work unit or task slug/id is provided, load it with get_work_unit / get_task; otherwise list candidates with list_work_units / list_tasks for selection.


AgiFlow Project Management Guidelines

Follow the shared AgiFlow project-management guidelines in references/agiflow-agents.md — agent assignment, the task status workflow and transitions, work-unit best practices, and the tags strategy apply to this workflow.


Steps

1. Load the Work for Review

If work unit slug/id provided:

  1. Use get_work_unit MCP tool to load the work unit and its tasks.
  2. For each task in the work unit, use get_task to load full details (acceptance criteria, devInfo, comments).

If task slug/id provided: 3. Use get_task MCP tool to load the task with full details.

If nothing provided: 4. Use list_work_units with status: "completed" to show completed work units. 5. Use list_tasks with status: "Review" or status: "Done" to show completed tasks. 6. Ask user to select what to review.

2. Acceptance Criteria Audit

  1. For each task, evaluate EVERY acceptance criterion:
CriterionVerdictEvidence
[criterion text]PASS / FAIL / PARTIAL[what you observed]

Verdict rules:

  • PASS: Criterion is fully met with evidence (test results, devInfo notes, implementation confirmed)
  • FAIL: Criterion is not met or cannot be verified
  • PARTIAL: Criterion is partially met — document what's missing
  1. Check devInfo for implementation evidence:
    • Were files actually changed? (filesChanged in devInfo)
    • Did tests pass? (testResults in devInfo)
    • Is there a draft PR? (draftPr in devInfo)
    • Were acceptance criteria marked as checked?

3. Quality Checks

  1. Evaluate these quality dimensions:

Completeness:

  • Are ALL acceptance criteria addressed (not just the easy ones)?
  • Are there gaps between what was asked and what was delivered?
  • Were test cases included for critical paths?

Consistency:

  • Do the changes align with the task/work unit description?
  • Is the scope appropriate (no gold-plating, no missing pieces)?
  • Do devInfo notes match the actual changes?

Risk Assessment:

  • Are there edge cases that weren't covered?
  • Could the changes break existing functionality?
  • Are there security implications?
  • Are there performance concerns?

4. Review Comments

  1. For each task reviewed, use list_task_comments to check existing progress notes.
  2. Use create_task_comment to add your review findings:
**Review Summary**

Verdict: APPROVED / NEEDS REWORK / APPROVED WITH NOTES

**Acceptance Criteria:**
- [criterion 1]: PASS - [evidence]
- [criterion 2]: FAIL - [what's missing]

**Issues Found:**
- [BLOCKER] [description] — must fix before shipping
- [WARNING] [description] — should fix, creates tech debt
- [NOTE] [description] — minor improvement opportunity

**What's Good:**
- [positive observation]

**Next Steps:**
- [action needed, if any]

5. Verdict and Actions

  1. Based on findings, recommend one of:

APPROVED — All criteria pass, no blockers.

  • If work unit: use update_work_unit to confirm status "completed"
  • If task: use update_task to confirm status "Done"
  • Summarize what shipped and its value

APPROVED WITH NOTES — All criteria pass, minor issues.

  • Keep current status
  • Create follow-up tasks for noted improvements using batch_create_tasks
  • Document notes in task comment

NEEDS REWORK — One or more criteria fail or blockers found.

  • If work unit: use update_work_unit to set status back to "in_progress"
  • If task: use update_task to set status back to "In Progress"
  • Uncheck failed acceptance criteria
  • Document what needs to change in task comment
  • Be specific: "criterion X fails because Y, to fix do Z"

6. Work Unit Summary (if reviewing a work unit)

  1. After reviewing all tasks in the work unit, produce a rollup:
**Work Unit Review: [title]**

Tasks: [N passed] / [total] passed
Overall: APPROVED / NEEDS REWORK

| Task | Slug | Verdict | Issues |
|------|------|---------|--------|
| [title] | [slug] | PASS | 0 |
| [title] | [slug] | FAIL | 2 blockers |

[Summary of what the work unit delivers and whether it's ready]

Common Mistakes to Avoid

  • Rubber-stamping: marking PASS without checking evidence
  • Fixing issues during review instead of documenting them for rework
  • Ignoring devInfo (it contains the implementation trail)
  • Reviewing only the happy path and missing edge cases
  • Not creating follow-up tasks for issues found
  • Approving work with unchecked acceptance criteria

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.