# Verification Gates > Enforces build, test, and lint gates with a self-correction loop and retry budget. Use when this capability is needed. - Skill: `tomevault-io/verification-gates-3` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/verification-gates-3` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/verification-gates-3/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Finance & Business - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/verification-gates-3 --- # Verification Gates & Self-Correction Loop To ensure repository stability, all major changes must pass through a series of "Verification Gates." ## The Gates 1. **Build Gate**: `npm run build` (or equivalent). 2. **Test Gate**: `npm test` (or equivalent). 3. **Lint Gate**: `npm run lint` or `ruff check`. ## The Self-Correction Workflow Whenever a gate fails (e.g., via the `gate` tool), follow this loop: 1. **Capture**: Inspect the `stderr` and `stdout` provided by the `gate` tool. 2. **Analyze**: Explicitly state the root cause of the failure. 3. **Fix**: Apply a patch or modification to address the cause. 4. **Retry**: Re-run the `gate` tool with the same command. ## Retry Budget - You have a **Retry Budget of 5 attempts**. - If the gate still fails after 5 tries, **STOP IMMEDIATELY** and ask the human user for intervention. Do not continue trying variations. ## Operational Guidance Always run the `gate` tool for any verification step. If a gate fails, announce "SELF-CORRECTION LOOP START [Attempt X/5]". --- > Converted and distributed by [TomeVault](https://tomevault.io/claim/icampana) — claim your Tome and manage your conversions.