agentrigs
+ submit rig
~ / rigs / Anyesh / skillprobe

Anyesh/skillprobe

Automated end-to-end testing for AI agent skills(agentskills.io). Launches Claude Code and Cursor as subprocesses, runs scenarios in real workspaces, and asserts what the model actually does.

↗ GitHub ★ 6 MIT updated 6mo ago project Claude Code
share on X
ARCHETYPE
Minimalist
Lean instructions, few tools, tiny context tax. Lets the model think.
CONTEXT TAX · EVERY TURN
~54 tokens
Featherweight · median rig: 2.3k · breakdown
GUARDRAILS
0/5
No committed guardrails · details

Copy this rig

# review before running: this installs third-party code
$ npx degit Anyesh/skillprobe/.claude ./rig-skillprobe  # inspect, then merge into .claude/

MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.

$ npx degit Anyesh/skillprobe/.agents/skills/write-tests .claude/skills/write-tests
$ npx degit Anyesh/skillprobe/examples/skills/bad-python .claude/skills/bad-python
$ npx degit Anyesh/skillprobe/examples/skills/clean-code .claude/skills/clean-code

This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:

{
  "permissions": {
    "deny": [
      "Read(./.env)",
      "Read(**/.env)",
      "Read(~/.ssh/**)",
      "Bash(rm -rf *)",
      "Read(**/*.pem)",
      "Bash(rm -rf /)",
      "Bash(git push --force:*)",
      "Bash(sudo *)",
      "Read(.env)",
      "Bash(rm -rf /*)",
      "Read(./.env.*)",
      "Read(~/.aws/**)",
      "Bash(git push --force*)",
      "Bash(rm -rf:*)",
      "Read(**/*.key)",
      "Read(**/.env.*)",
      "Bash(sudo:*)",
      "Bash(git reset --hard*)",
      "Bash(git reset --hard:*)",
      "Read(.env.*)"
    ],
    "ask": [
      "Bash(git push:*)",
      "Bash(git push *)",
      "Bash(git commit:*)",
      "Bash(rm *)",
      "Bash(rm:*)",
      "Bash(git rebase *)",
      "Bash(wget *)",
      "Bash(npm publish:*)",
      "Bash(git commit *)",
      "Bash(gh pr merge *)"
    ]
  }
}

Skills (3)

write-testsbad-pythonclean-code

Similar rigs

karawanshy/kiro2claude
Use .kiro/ agents, skills, steering, specs and hooks directly from Claude Code — one source of truth, no duplicated AI config.
★ 0 project Claude Code 1 mcp 1 skills 6d ago
Minimalist 482 tok ·
JuanJo24S/claude-develop-skills
Skills and a git-guard hook for Claude Code and OpenCode: branch and PR workflow, secrets out of the code, clean code. Docs in Spanish and English.
★ 0 collection Claude Code 4 skills 8d ago
Minimalist 292 tok ·
azharuddinkhan8898/ai-core
Centralized AI registry and distribution system for reusable AI assets — skills, instructions, agents, and MCP configs — across coding agents like GitHub Copilot, Claude Code, Cursor, and Windsurf.
★ 0 project Claude Code 7 skills 4mo ago
Pragmatist 408 tok ·
yousefkadah/groundrules
One source of truth for AI coding agents — detect your stack, scaffold security-first standards + skills, and generate AGENTS.md / CLAUDE.md / Cursor / Copilot / Gemini from one place. Multi-stack.
★ 1 project Claude Code 8 skills 3mo ago
Pragmatist 623 tok ·
copied ✓