001-review

Reviews code changes and documents against XDRS Policies and reports structured compliance findings.

1|1|Updated Feb 28, 2026
One-click install
npx skills add https://github.com/flaviostutz/xdrs-core --skill 001-review-flaviostutz
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: 001-review
Source: https://github.com/flaviostutz/xdrs-core/tree/main/examples/typed-scope/.xdrs/_core/adrs/principles/skills/001-review
Command: npx skills add https://github.com/flaviostutz/xdrs-core --skill 001-review-flaviostutz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @mermaid-js/mermaid-cli.

What problem does it solve? Teams accumulate many decision Policies across scopes, and manually checking whether code changes or documents comply with them is slow and error-prone. This Skill automates that review, producing structured findings with severity, Policy references, and fixes. ## Core Features & Use Cases - Policy-aware review: Compiles applicable Policies from hierarchical XDRS scopes, resolving follows, extends, scope-type chains, and local meta-policies before reviewing. - Structured findings: Classifies violations as ERROR or WARNING with file/line references, Policy citations, and actionable fixes, ending in a PASS/FAIL outcome. - Diagram validation: Extracts fenced Mermaid blocks and validates them with mermaid-cli, reporting render errors. - Use Case: After opening a pull request, ask for a compliance review; the Skill diffs against the base branch, loads all governing Policies, and returns a findings report without modifying any code. ## Quick Start Review my current git diff against the XDRS Policies in this repository and report any violations.

Frequently Asked Questions about 001-review

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

FAQPage Schema
How do I review code changes against architecture decision records?▼

Run this Skill on a git diff or file set; it compiles all applicable Policies from the XDRS index, cross-references each change, and reports ERROR or WARNING findings with Policy references and suggested fixes.

How does the review handle nested or inherited policy scopes?▼

The Skill reads each scope's index.md for follows and extends fields, loads inherited policies depth-first, resolves scope-type chains, and fails the review if any declared governance layer is missing or unreadable.

Can this Skill validate Mermaid diagrams in markdown files?▼

Yes. It extracts every fenced mermaid code block, writes it to a temporary file, and runs mermaid-cli validation. Diagrams that fail are reported as ERROR findings with file path and error message.

Does the review modify my code automatically?▼

No. The Skill is read-only by design. It reports findings with specific fix suggestions but never edits code; you must request changes separately.

Why did my review fail before producing any findings?▼

The prerequisites gate fails when a scope listed in follows or extends is missing, a local meta-policy is unreadable, or no scope-type policy exists. Install the missing governance scopes and rerun the review.