dev-docs-review

Runs multi-agent code review of a completed implementation phase via a workflow script.

16|1|Updated May 5, 2026
One-click install
npx skills add https://github.com/AIBiz-Automatyzacje/workspace-template-mobile --skill dev-docs-review-aibiz-automatyzacje
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: dev-docs-review
Source: https://github.com/AIBiz-Automatyzacje/workspace-template-mobile/tree/main/.claude/skills/dev-docs-review
Command: npx skills add https://github.com/AIBiz-Automatyzacje/workspace-template-mobile --skill dev-docs-review-aibiz-automatyzacje

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After implementing a phase of a task, teams need a structured, multi-perspective code review that checks security, spec compliance, simplicity, and test coverage, plus E2E verification, without manually coordinating reviewers. ## Core Features & Use Cases - Multi-agent review orchestration: Launches the dev-docs-review-wf.js workflow which routes findings to domain reviewers (security, spec-compliance, simplicity, test-coverage, plus conditional performance, architecture, typescript, and e2e reviewers). - Severity gate decision: Consolidates findings into P1/P2/P3 counters and produces a BLOKUJE / ZASTRZEŻENIA / CZYSTE verdict that decides whether work can continue. - Checkbox bookkeeping: Automatically resolves Weryfikacja: and Test: [E2E] checkboxes by running CLI commands, greps, or matching Maestro E2E results, and writes a review-faza-N.md report. - Use Case: After finishing phase 2 of a task documented in docs/active/my-task/, invoke the review to get a full report with plan-deviation analysis, E2E Maestro results, and an updated task file with fix checkboxes. ## Quick Start Ask the assistant to run dev-docs-review on the task folder docs/active/my-task for phase 2 and summarize the severity gate result.

Frequently Asked Questions about dev-docs-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a multi-agent code review on a finished implementation phase?▼

Invoke the skill with the task folder path and phase number, for example the folder under docs/active and phase 2. It launches the dev-docs-review-wf.js workflow which routes findings to domain reviewers and returns a consolidated severity-gated report.

What reviewers participate in the automated code review?▼

The core reviewers are security, spec-compliance, simplicity, and test-coverage, which always run. Performance, architecture, typescript, and e2e reviewers activate only when their domain is present in the phase, with fail-open full routing when no layer flags exist.

How does the severity gate decide if work can continue?▼

Findings are classified as P1 blocking, P2 important, or P3 nit. Any P1 blocks continuation, only P2 means continue with reservations, and only P3 means ready to proceed with optional suggestions.

Can the review automatically check off verification and E2E test checkboxes?▼

Yes. CLI and grep-style checkboxes are executed via Bash and checked off on exit code 0. E2E checkboxes marked [E2E] are resolved only from the Maestro tester's run results, with FAIL or SKIP entries left unchecked and annotated.

What happens when the E2E tester crashes during review?▼

If the tester crashed, no E2E checkboxes are modified and no suffixes are added. The orchestrator stops the run and the review repeats with the tester restored, so results are never inferred from missing evidence.