regression-check

Classify residual regression risk from diffs and verification reports.

Updated Apr 9, 2026
One-click install
npx skills add https://github.com/ldmrepo/agent-harness --skill regression-check-ldmrepo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: regression-check
Source: https://github.com/ldmrepo/agent-harness/tree/main/template/.claude/skills/regression-check
Command: npx skills add https://github.com/ldmrepo/agent-harness --skill regression-check-ldmrepo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill standardizes post-change regression assessment to prevent narrow fixes or bugfixes from breaking adjacent behavior, hiding side effects, or degrading repository health without explicit tracking.

Core Features & Use Cases

  • Evidence-driven assessment: maps changed surface to likely regression zones and aggregates verification evidence such as diffs, changed tests, smoke, and full verification reports.
  • Policy-aware gating: classifies residual risk against configured thresholds and influences approval outcomes by setting approve/pending/reject actions and follow-up tasks.
  • Artifact updates and traceability: records findings in session summary, known issues, current task, and backlog when follow-up work is required.
  • Use Case: a reviewer validating a bugfix or feature that touched shared runtime or API contracts and needing a narrow, reproducible regression verdict.

Quick Start

Use the regression-check skill to analyze the recent change, run focused verification for adjacent areas, classify residual regression risk, and record findings in the session summary.

Frequently Asked Questions about regression-check

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

FAQPage Schema
How do I assess regression risk after applying a bugfix to shared runtime contracts?

Assess regression risk after a bugfix by mapping the changed surface to likely regression zones and aggregating verification evidence like diffs, smoke reports, and changed tests. This classifies residual risk against configured thresholds to produce a reproducible regression verdict.

What is evidence-driven regression validation for bounded implementations?

Evidence-driven regression validation maps changed surface areas to likely regression zones and aggregates verification evidence such as diffs, changed tests, smoke reports, and full verification reports to classify residual regression risk with supporting proof.

Can I use policy-aware gating to block code review approval for high residual regression risk?

Policy-aware gating classifies residual regression risk against configured thresholds to influence code review approval outcomes. It explicitly sets approve, pending, or reject actions and generates follow-up tasks when residual risk exceeds acceptable levels.

How do I track residual regression findings in session artifacts and known issues?

Track residual regression findings by recording the classified risk evidence in session summaries, known issues, current tasks, and backlogs. This artifact update mechanism ensures traceability and schedules follow-up work when verification reports indicate lingering side effects.

What do I need to provide to run a targeted regression check on recent code changes?

Running a targeted regression check requires access to session artifacts and repository state, including diffs, changed tests, smoke reports, and full verification reports. The skill analyzes these inputs to produce classified residual risk with evidence.