← Back to Skills

Sherlock Review

🌐Community
by proffesor-for-testing · vlatest · Repository

Analyzes text data to identify inconsistencies, biases, and potential errors like Sherlock Holmes investigates a case.

Install on your platform

We auto-selected Claude Code based on this skill’s supported platforms.

1

Run in terminal (recommended)

terminal
claude mcp add sherlock-review npx -- -y @trustedskills/sherlock-review
2

Or manually add to ~/.claude/settings.json

~/.claude/settings.json
{
  "mcpServers": {
    "sherlock-review": {
      "command": "npx",
      "args": [
        "-y",
        "@trustedskills/sherlock-review"
      ]
    }
  }
}

Requires Claude Code (claude CLI). Run claude --version to verify your install.

About This Skill

What it does

The Sherlock Review skill helps AI agents analyze code claims and related data to determine their validity. It uses a structured investigative process, inspired by Sherlock Holmes, to gather evidence (code changes, tests, behavior), deduce whether reality matches the claim, and ultimately conclude with a supported verdict. The tool focuses on verifying claims made in pull requests or commit messages through rigorous examination of code and testing results.

When to use it

  • Verifying claims made in pull request descriptions or commit messages.
  • Assessing the accuracy of reported bug fixes or new feature implementations.
  • Determining whether code changes are fully supported by accompanying tests.
  • Investigating discrepancies between expected behavior and actual runtime output.
  • Analyzing code contributions to ensure they align with stated goals.

Key capabilities

  • Evidence Gathering: Collects data like git diff outputs, test coverage reports, and observed behavior.
  • Deductive Analysis: Compares claims against gathered evidence to identify inconsistencies.
  • Verdict Levels: Categorizes findings as TRUE, PARTIALLY TRUE, FALSE, or NONSENSICAL.
  • Investigation Template: Provides a structured format for documenting the investigation process and results.
  • Holmesian Principles: Applies principles like prioritizing data collection and eliminating impossible scenarios.

Example prompts

  • "Analyze this pull request claim: 'Fixes issue #123 by updating the login function.'"
  • "Verify if the commit message 'Adds new feature X' is supported by the code changes and tests."
  • "Investigate the claim that bug Y has been resolved in this branch."

Tips & gotchas

  • The skill relies on access to Git repositories and testing frameworks. Ensure the agent has appropriate permissions and environment setup.
  • The "Evidence Collection Checklist" provides guidance on what data to gather for a thorough investigation.
  • Focus on reproducible evidence; avoid assumptions or subjective interpretations.
View Repository →

Tags

🛡️

TrustedSkills Verification

Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.

Security Audits

Gen Agent Trust HubPass
SocketPass
SnykPass

Details

Version
vlatest
License
Author
proffesor-for-testing
Installs
33
Repository (canonical source) →

🌐 Community

Passed automated security scans.