agentleFS
Codagent-AICodagent-AI/agent-skills/skills/review-approach/SKILL.md

review-approach

Skill31 starsChanged 6 months agoMITView on GitHub

Performs the final review of a completed proposal, product specification, technical design, and test plan for cross-artifact consistency, consequential gaps, missing decisions, weak tradeoffs, failure modes, testing strategy, and better alternatives; use when asked to review an approach, challenge design or testing decisions, find gaps in completed definition artifacts, or provide a second opinion before task planning or implementation.

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read every definition artifact and relevant repository instructions.

    line 11

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “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

Review Approach

Review the completed proposal, specifications, design, and test plan together before task planning. Determine whether they form a coherent, implementation-ready definition with sound behavioral, technical, and testing decisions.

Read every definition artifact and relevant repository instructions. Inspect affected code, tests, and established patterns enough to judge feasibility and fit. Treat omissions as findings when an implementer or tester would otherwise have to invent a consequential decision.

Review focus

Trace important user flows and system interactions end to end. Look for material:

  • contradictions, scope drift, dropped commitments, terminology drift, or incompatible assumptions;
  • missing or untestable behavior, boundaries, state transitions, failure handling, permissions, data lifecycle, compatibility, or migration decisions;
  • unclear ownership, component boundaries, data flow, concurrency, recovery, security, operability, rollout, or observability;
  • testing strategy that conflicts with the definition, misses important integration or critical journeys, uses the wrong test layer, leaves acceptance substitutes ambiguous, or relies on unstated human judgment;
  • avoidable coupling, complexity, maintenance burden, irreversible commitments, weak rationale, or overlooked alternatives.

Apply judgment rather than mechanically filling a checklist. Do not review task decomposition, implementation code quality, formatting, or parser mechanics except where they prevent the definition from being usable. Do not relitigate whether the approved feature should exist, broaden its scope, or invent optional features.

Report

For each consequential finding, cite the exact artifact section and repository evidence, classify it as a cross-artifact inconsistency, missing decision, or challenged decision, explain the concrete risk, and recommend a resolution or small set of real alternatives with a preferred choice.

Rank findings by impact and end with a direct readiness assessment. Say explicitly when the approach is sound and no consequential gap remains. Do not edit artifacts.

## Findings

### [high|medium|low] <title>
- Artifacts: <exact sections and repository evidence>
- Type: <inconsistency | missing decision | challenged decision>
- Risk: <concrete consequence>
- Recommendation: <preferred resolution and material alternatives>

## Open Questions
<Only decisions needed to resolve findings, or "None.">

## Overall Assessment
<Whether the definition is ready for task planning and why.>

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.