AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified Apache-2.0 Self-run

Review Prod Ready

skill-sakost-review-prod-ready-skill-review-prod-ready-skill · by sakost

Use when comprehensive production-readiness code review is needed after implementation — dispatches up to 7 specialized parallel reviewers covering tests, plan completeness, architecture, complexity, DRY violations, suppressed warnings, and production readiness

No reviews yet
0 installs
19 views
0.0% view→install

Install

$ agentstack add skill-sakost-review-prod-ready-skill-review-prod-ready-skill

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-sakost-review-prod-ready-skill-review-prod-ready-skill)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
3mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Review Prod Ready? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Production-Ready Code Review

Dispatches up to 7 specialized review subagents in parallel. Each reviewer focuses on ONE dimension deeply, catching issues a single broad reviewer misses.

Git Context (auto-detected)

HEAD: !git rev-parse --short HEAD 2>/dev/null || echo "not a git repo" Base: !git merge-base HEAD main 2>/dev/null || git merge-base HEAD master 2>/dev/null || echo "no base found" Changed files: !git diff --stat $(git merge-base HEAD main 2>/dev/null || git merge-base HEAD master 2>/dev/null) HEAD 2>/dev/null || echo "run git diff manually"

Process

1. Read reviewer prompts

Read ${CLAUDE_SKILL_DIR}/reviewers.md for the detailed prompt for each of the 7 reviewers.

2. Dispatch reviewers in parallel

Dispatch ALL applicable reviewers simultaneously using the Agent tool. Pass each reviewer:

  • The git range from the context above
  • The list of changed files above
  • Their specific prompt from reviewers.md
  • User context if provided: $ARGUMENTS

Each agent should:

  • Read full content of changed files (not just the diff — bugs hide in unchanged lines nearby)
  • Focus ONLY on their assigned dimension
  • Output findings in the structured format below

Skip reviewers that don't apply (e.g., skip Plan Completeness if no plan exists, reduce count for small diffs <5 files).

3. Aggregate results

Aggregation rules — do NOT soften findings:

  • Keep every reviewer's severity label EXACTLY as they assigned it. Do not downgrade Critical → Important to "seem balanced."
  • Do not drop findings to shorten the report. If two reviewers flagged overlapping issues, merge them but preserve the HIGHER severity.
  • Do not rewrite "what's wrong" descriptions to sound more polite. Keep the blunt language from the reviewer.
  • The verdict at the bottom must reflect the findings, not a diplomatic compromise. If ANY Critical issue exists, verdict is No or With fixes — never Yes.
  • If you feel tempted to soften a finding, that's a sign the reviewer was right and the author won't like it. Keep it.

After all reviewers complete, combine into a single report:

## Production Readiness Review

**Range:** BASE..HEAD | **Files:** N changed | **Date:** YYYY-MM-DD

### Critical Issues (must fix before merge)
[Combined from all reviewers, deduplicated, with source reviewer noted]

### Important Issues (should fix)
[Combined, deduplicated]

### Minor Issues (nice to have)
[Combined, deduplicated]

### Suppression Audit
| File:Line | Directive | Verdict | Reason |
|-----------|-----------|---------|--------|

### Verdict
**Ready to merge?** Yes / No / With fixes
**Top risks if merged as-is:**
1. ...
2. ...
3. ...

Issue Format (for each reviewer)

Each finding must include:

  • Severity: Critical / Important / Minor
  • File:line reference
  • What's wrong (specific, not vague)
  • Why it matters (impact if not fixed)
  • Suggested fix (if not obvious)

Source & license

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

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

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.