- Home
- Skills
- Yuniorglez
- Gemini Elite Core
- Artifact Janitor
artifact-janitor_skill
- Python
7
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill yuniorglez/gemini-elite-core --skill artifact-janitor- SKILL.md5.2 KB
Overview
This skill is a tactical cleanup and optimization tool for modern 2026 development environments. It reclaims disk space, removes corrupted build artifacts, and restores build health in large monorepos and JS/TS workspaces. The focus is safe, automated discovery and removal of caches and redundant dependencies without risking repository integrity.
How this skill works
artifact-janitor scans projects and user caches to locate multi-gigabyte build artifacts (node_modules, .next, .turbo, tsbuildinfo, framework caches). It offers dry-run analysis, targeted removal, and automation via a robust deep-clean.sh script that reports reclaimed space and enforces safety checks. Post-clean verification steps ensure builds and type checks succeed before finishing.
When to use it
- When disk usage from build caches exceeds acceptable limits or slows developer workflows
- If builds show intermittent or "ghost" errors that likely come from corrupted artifacts
- Before archiving or sharing a repository to avoid shipping local artifacts
- During CI tuning to remove unnecessary caches while preserving incremental performance
- When onboarding contributors to quickly normalize environment sizes across machines
Best practices
- Always run --dry-run first to preview deletions and sizes
- Protect vital files like .git and .env by adding them to a protected list
- Check for running node/bun/vite processes before deleting to avoid partial cleanup
- Prefer targeted invalidation (tsconfig.tsbuildinfo, framework caches) over blind rm -rf
- Use automation hooks in CI with conservative settings to avoid breaking incremental builds
Example use cases
- Resolve "Property X does not exist" type errors by removing tsconfig.tsbuildinfo and .next types
- Reclaim dozens of GBs in a monorepo by removing recursive node_modules and framework caches
- Pre-archive scrub: run deep-clean.sh --dry-run then execute to create a lean project artifact
- CI cleanup job that prunes pnpm store and workspace node_modules with safety checks
- Recover from branch-switch build failures by wiping node_modules and reinstalling with bun
FAQ
Yes if you use --dry-run first and confirm; the tool enforces safety checks and protects .git/.env when configured.
Will cleaning break incremental CI caches?
If used blindly, yes. Prefer targeted invalidation and configure CI to only prune caches that are known-bad or unnecessary for increments.