RE

release-candidate-gate

Validates code against security and regression benchmarks without triggering deployments.

Install

mkdir -p .claude/skills/release-candidate-gate && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/12816" && unzip -o skill.zip -d .claude/skills/release-candidate-gate && rm skill.zip

Installs to .claude/skills/release-candidate-gate

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Run the non-deploying Agent Red release-candidate gate before treating a build as production deployable. Covers Python security checks, targeted regression tests, frontend builds, and GroundTruth governance adoption checks.
223 charsno explicit “when” trigger
Intermediate

Key capabilities

  • Run Python security checks
  • Execute targeted regression tests
  • Perform frontend builds
  • Check GroundTruth governance adoption
  • Report GO/NO-GO confidence
  • Record required evidence for release

How it works

The skill executes a Python script with specified options to perform various checks, and reports GO/NO-GO confidence based on the outcomes of these checks.

Inputs & outputs

You give it
Build to be evaluated for production readiness, with optional arguments like --python-only, --frontend-only, or --full
You get back
GO/NO-GO confidence report and recorded evidence

When to use release-candidate-gate

  • Validate production readiness
  • Run security scans
  • Check regression tests

About this skill

<!-- GTKB-CODEX-SKILL-ADAPTER Generated: true Generated by: scripts/generate_codex_skill_adapters.py Canonical source: .claude/skills/release-candidate-gate/SKILL.md Canonical source sha256: 149f282f610457a732f382dd16d8853e421c6585b5036a99ee2cc3567b770a92 Generated at: 2026-05-06T18:52:34Z Do not edit this adapter directly. Edit the canonical source and regenerate. GTKB-CODEX-SKILL-ADAPTER -->

Release Candidate Gate

Use this skill before calling a build production-ready. The gate is non-deploying: it does not push images, update Azure resources, call live services, or mutate external infrastructure.

Commands

Run the complete local gate:

python scripts/release_candidate_gate.py --include-frontend

Run the Python/security side only:

python scripts/release_candidate_gate.py --skip-frontend

Run the frontend side only:

python scripts/release_candidate_gate.py --skip-python --include-frontend

CI must prove the Python gate under the production target interpreter:

python scripts/release_candidate_gate.py --require-python 3.12 --skip-frontend

Required Evidence

Before reporting GO confidence, record:

  • release gate command, result, and key counts;
  • any skipped live/deploy checks and why they are out of scope;
  • MemBase update path or KnowledgeDB document ID;
  • Deliberation Archive source reference or DELIB ID;
  • regression test files that would fail if the change regressed.

Stop Conditions

Stop and report NO-GO if:

  • the generated production gateway manifest exists or is still tracked;
  • Ruff E/F, import-cycle, Bandit, or pip-audit gates fail;
  • targeted auth/config/deploy/release-gate tests fail;
  • frontend tests or any admin/widget build fails;
  • GroundTruth-KB governance adoption checks fail;
  • Python 3.12 proof is missing for the build being considered for deployment.

When not to use it

  • When the build is already production-ready without prior checks
  • When the generated production gateway manifest exists or is still tracked
  • When Ruff E/F, import-cycle, Bandit, or pip-audit gates fail

Limitations

  • Does not push images or update Azure resources
  • Does not call live services or mutate external infrastructure
  • Requires Python 3.12 proof for the build being considered for deployment

How it compares

This skill automates a series of pre-deployment checks and evidence collection, providing a structured gate before a build is considered production-ready, unlike manual, ad-hoc testing.

Compared to similar skills

release-candidate-gate side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
release-candidate-gate (this skill)03moNo flagsIntermediate
verification-loop04moReviewIntermediate
ci02moReviewIntermediate
dependency-auditor19moReviewBeginner

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry