Skip to main content
skill CX Codex

adjudicate-regression — Claude Code skill: setup and use

adjudicate-regression — a skill for Claude Code and Codex that reviews reproduction evidence to confirm whether a flaw is a true regression or an exclusion. It evaluates live artifacts before logging issues.

CC-BY-SA-4.0 669 installs ★ 8,341 updated 2026-10-24

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add trailofbits/skills --skill adjudicate-regression -a claude-code
Open on GitHub

What adjudicate-regression does

The skill inspects frozen source snapshots, reproduction setups, PoCs, and raw execution logs. It cross-checks baseline S0 against modified S1 through active interfaces. An issue exposing an existing defect or an unfinished fix gets ruled out from regression counts.

Evaluation runs in an isolated context away from reproduction and replaying agents. If input proofs show missing steps, the agent refuses to synthesize missing evidence and reports gaps back to the caller instead.

Every decision seals into decision.json with citations for each finding. When evidence remains inconclusive, the output records an UNRESOLVED state alongside the exact unverified checks.

What it helps with

  • distinguish fresh regressions from latent pre-existing defects
  • rule out incomplete fixes and intentional interface updates
  • verify matched PoC execution on live S0 and S1 states
  • record auditable findings inside sealed decision.json files
  • deduplicate incident candidates without confirming unproved bugs

Example prompts for the agent

  • Evaluate the reproduction package for bug-18 and produce decision.json
  • Check whether this live PoC failure constitutes a regression or an exclusion
  • Inspect the S0 and S1 logs to see if the patch introduced this defect

How to install adjudicate-regression

1Skills CLI (Vercel)

npx skills add trailofbits/skills --skill adjudicate-regression -a claude-code

2Manually: put the files in place

~/.claude/skills/adjudicate-regression/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/adjudicate-regression/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about adjudicate-regression

How to install adjudicate-regression in Claude Code?

Run npx skills add trailofbits/skills --skill adjudicate-regression -a claude-code. The file will be placed in ~/.claude/skills/adjudicate-regression/SKILL.md.

What does adjudicate-regression do?

It examines candidate defect logs and reproduction runs to determine if a change caused a new regression or an excluded defect.

What is required for adjudicate-regression to work?

Frozen source trees, reproduction packages, verified run logs across S0/S1 states, and a fresh agent context.

Similar skills

All skills for Testing and QA →

From the same set95

The whole set →
prove-patch prove-patch — a skill for Claude Code validates a local security patch with scoped cases and a three-arm comparison. It runs when a patch commit needs independent proof. CX Codex +1 676installs hunt-regressions hunt-regressions — a skill for Claude Code that discovers potential regressions before reproduction. Runs when analyzing code changes and their callers. CX Codex +1 675installs reproduce-regression The reproduce-regression Claude Code skill reproduces a regression candidate on live systems before and after a change. It builds a PoC and compares both versions using the same input. CX Codex +1 671installs report-regressions report-regressions — a skill for Claude Code builds regression reports from saved campaign decisions and links findings to evidence. It runs when a campaign report is created or refreshed. CX Codex +1 670installs prepare-regression prepare-regression — a skill for Claude Code that pins source revisions and gathers commit or PR context for a regression investigation. It runs during investigation setup. CX Codex +1 670installs replay-regression The replay-regression Claude Code skill independently tests a supplied PoC against fresh builds from before and after a change. It runs the same input on S0 and S1, then records what happened. CX Codex +1 668installs cvss-scoring The cvss-scoring Claude Code skill turns a vulnerability description into a CVSS v4.0 vector. It activates for requests to score a finding or estimate its severity. CX Codex +1 304installs taint-trace The taint-trace Claude Code skill adjusts a project's static taint-analysis models, then runs its analyzer. It addresses missed data paths, unresolved framework calls, and false positives caused by imprecise models. CX Codex +1 130installs

Project sponsors

Companies that keep this analytics open