verification-quality

Verify code quality and agent outputs with truth scoring and automatic rollback.

15|1|Updated Apr 26, 2026
One-click install
npx skills add https://github.com/monoes/monomind --skill verification-quality-monoes
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verification-quality
Source: https://github.com/monoes/monomind/tree/main/.claude/skills/verification-quality
Command: npx skills add https://github.com/monoes/monomind --skill verification-quality-monoes

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams catch low-quality code, unreliable agent output, and failing changes before they spread, so verification becomes automatic instead of manual.

Core Features & Use Cases

  • Truth Scoring: Track reliability from 0.0 to 1.0 for code, agents, and tasks.
  • Verification Checks: Validate correctness, best practices, security, performance, and documentation.
  • Automatic Rollback: Revert failed changes to the last known good state, optionally selectively.
  • Reports and Dashboards: Generate JSON, CSV, HTML, and live dashboard views for monitoring and CI/CD.
  • Use Case: A developer runs verification on a feature branch, reviews the score, and lets the skill roll back unsafe edits before merging.

Quick Start

Ask the skill to run verification on your current project, explain the score, and roll back any failed changes below the configured threshold.

Frequently Asked Questions about verification-quality

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

FAQPage Schema
How do I automatically roll back failed code changes in a CI pipeline?

Automatic rollback reverts failed changes to the last known good state using git-backed recovery when verification scores fall below configured thresholds. It selectively reverts unsafe edits in files or directories before merging.

What is truth scoring for code quality and how does it work?

Truth scoring tracks reliability on a 0.0 to 1.0 scale for code, agents, and tasks. It validates correctness, security, performance, and documentation to prevent low-confidence changes from persisting in development workflows.

How do I run verification checks on a feature branch before merging?

Run verification on your current project to review the reliability score and automatically roll back unsafe edits. The skill applies threshold-based checks to files and directories to catch failing changes early.

Can I export verification reports as JSON or HTML for dashboards?

Verification reports export as JSON, CSV, and HTML formats. You can generate these alongside live dashboard views for monitoring development workflows and CI/CD pipeline status.

Does watch mode support continuous verification for monitored development workflows?

Watch mode continuously monitors files, directories, and tasks to verify code quality and agent outputs. It applies threshold-based checks and triggers git-backed automatic recovery whenever verification fails.

What is the best way to prevent low-confidence agent outputs from persisting?

Prevent low-confidence outputs by applying threshold-based verification checks to tasks and agents. The skill scores reliability from 0.0 to 1.0 and rolls back changes failing the configured threshold.