Check Skills — Development skill for Claude Code
Vendor-neutral CLI for validating portable Agent Skills.
How to install Check Skills
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open spences10/check-skills and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Check Skills does
Vendor-neutral CLI for validating portable Agent Skills.
Alternatives in Development
- Validate Plan — You are tasked with validating that an implementation plan was correctly executed, verifying all success crite 10k ★
- Bypass 403 — Probe a 403/401 endpoint with the most-paid bypass tricks (header injection, path encoding, method swap, WAF f 4.5k ★
- Agentic Stack — One brain, many harnesses 2.2k ★
README
check-skills
[](https://viteplus.dev) [](https://vitest.dev)
Vendor-neutral CLI for validating, linting, and improving portable Agent Skills that follow the [agentskills.io specification](https://agentskills.io/specification).
Usage
`check-skills` is designed to be used inline in LLM coding sessions. After an agent creates or edits a skill, tell it to validate that skill:
Create a portable Agent Skill for , then run:
pnpx check-skills validate
For repositories containing many skills, tell the agent to validate the repo recursively:
Validate all skills in this repo with:
pnpx check-skills validate . --recursive --llm
Recommended inline commands:
# After creating or editing one skill
pnpx check-skills validate ./my-skill --llm
# After editing a skills repository
pnpx check-skills validate . --recursive --llm
# Machine-readable output for automation or agent parsing
pnpx check-skills validate . --recursive --json
# CI/spec-only validation against agentskills.io requirements
pnpx check-skills validate . --recursive --no-quality
Fix all errors before finishing. Treat warnings as quality prompts during authoring; use `--no-quality` when you only want spec compliance.
Commands
`validate `
Validates skill directories.
Options:
--recursive— discover skills recursively--strict— treat warnings as failures--json— machine-readable output--llm— concise stable output for coding agents--quiet— only show skills with problems--format github— emit GitHub workflow annotations--agent— run optional adapter checks (codex,claude-code,opencode,cursor,windsurf,pi)--no-quality— only run spec compliance checks
`explain <
Related Skills
IDEA.Md
IDEA.md — A standard file for portable idea intent in agentic software work. Share the idea, not the code. You
Agent Context Contract Article Demo
ContextContract: scan a CLAUDE.md into portable, vendor-bound and enforce-not-describe lines; compile AGENTS.m
Portability
Check that the governance configs stay consistent across every supported host, and emit a host-neutral CI snip
Tmux Rc
Watch and control whatever runs in a tmux pane — Codex, Claude Code, Gemini CLI, OpenCode, or any program that
Mine Gemini Workflows
Find repeated workflows in selected Gemini CLI sessions and draft portable skills after review.
Claude Config Portable
Portable CLI to export and import your entire Claude Code setup (plugins, skills, MCPs, settings) as a single
Related Agents
Design Handoff
Agent Handoff Designer on the Atlas bench. Keeps real MCP and CLI next actions visible, state-bound, portable,
Sous Chef
Use this agent when reviewing Java/Spring Boot code, checking best practices, validating test coverage, or ens
Vnv Sprint Prep Agent.Agent
Runs the V&V sprint workflow — platform-parity check, cloning dev sprint stories to the V&V board, validating