context-review

Scan MEMORY.md for credential patterns and verify schema field names in CLAUDE.md.

3|Updated Mar 17, 2026
One-click install
npx skills add https://github.com/marcoguillermaz/claude-dev-kit --skill context-review-marcoguillermaz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: context-review
Source: https://github.com/marcoguillermaz/claude-dev-kit/tree/main/packages/cli/templates/tier-l/.claude/skills/context-review
Command: npx skills add https://github.com/marcoguillermaz/claude-dev-kit --skill context-review-marcoguillermaz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates a focused security and code-quality review of the project context by performing three checks: credential pattern detection, unresolved placeholder identification, and schema-name freshness validation.

Core Features & Use Cases

  • Credential verification: scans MEMORY.md for sensitive tokens and flags real values.
  • Placeholder detection: locates unresolved placeholders in CLAUDE.md and pipeline rules.
  • Schema consistency: verifies recent migrations or schema definitions against mentions in CLAUDE.md to prevent drift.
  • Use Case: Before releasing, run this skill to ensure there are no leaked credentials, placeholders, or stale field names in your docs.

Quick Start

Run the context-review skill to perform C1-C3 checks on your repository and return a structured report.

Frequently Asked Questions about context-review

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

FAQPage Schema
How do I scan project documentation for leaked credentials and unresolved placeholders?

To scan for credential leaks and unresolved placeholders, run an automated context review that greps MEMORY.md for real token values and flags non-code-block placeholder occurrences in CLAUDE.md and pipeline rules.

What is schema drift and how do I check for stale field names in my documentation?

Schema drift occurs when documentation mentions outdated database fields. You can check for it by verifying that field names in CLAUDE.md match the most recent schema migration or definition file found in the repository.

Can I audit CLAUDE.md and MEMORY.md for security issues before a release?

Yes, you can audit these files before releasing by running a focused security review that detects leaked credentials in MEMORY.md and unresolved placeholders in CLAUDE.md, returning a structured report.

What happens during a schema consistency check if no migration files exist?

During a schema consistency check, if no schema migration or definition files are located in the repository, the schema freshness validation step is marked as N/A and skipped in the final report.

How do I find unresolved placeholders in pipeline rules without false positives from code blocks?

To find unresolved placeholders while avoiding false positives, the review process scans .claude/rules/pipeline.md and CLAUDE.md, specifically reporting only non-code-block occurrences of placeholder patterns.