skills
Agent skills and an AGENTS.md workflow template — isolate in worktrees, build to a service layer, prove with evidence, ship with before/after proof and Greptile review loops. For Claude Code, Cursor, and Codex.
No outside reports on file
This tool is a workflow helper, not a security layer—it requires careful vetting of each skill, strong access controls on services it creates, and monitoring to prevent data leaks from agent outputs.
Based on public adoption and transparency signals as of Oct 22, 2026. Our opinion, not a code audit or security guarantee.
What it can access
- Local filesNo local folder access notedLOW
- Command executionNo shell command runner notedLOW
- Network reachConnects to outside service APINOTED
- CredentialsNo API token requiredLOW
- Package sourceGitHub (michaelshimeles/skills)PUBLIC
Inferred from this listing’s public package details and description. Check the project’s source before installing.
Install
No npm or PyPI package is listed. Install from source: github.com/michaelshimeles/skills ↗
About Skills
This is a template and collection of reusable skills (functions and workflows) designed to help AI agents like Claude, Cursor, and Codex perform tasks more reliably. It provides a structure for organizing agent capabilities, testing them with evidence, and deploying them as services.
Who it’s for
Developers building AI agent systems who want to share skills across projects, test agent outputs carefully, and integrate with code editors or AI tools. Teams using Claude or Cursor for code generation and automation.
Alternatives
- LangChain tools and agents — Framework for building LLM tools with built-in safety patterns.
- Semantic Kernel (Microsoft) — Structured skill composition for AI agents with validation.
- Native Claude API with tools — Anthropic's tool-calling feature without additional abstractions.
- Agent frameworks like AutoGPT or BabyAGI — Full agent orchestration (may have their own security gaps).
Setup
Check these steps against the project’s own instructions before running anything.
Follow the project's README at https://github.com/michaelshimeles/skills (opens external site) for installation and configuration details specific to your AI tool (Claude Desktop, Cursor, or Codex).
Why this grade
Signal-based grade (3/7 points): well starred on GitHub, public source code. This reflects adoption and transparency, not a code review.
Popularity and upkeep
- GitHub stars
- 1.4k
- Listing last refreshed
- 2026-10-25
These are public popularity signals, not a maintenance check. Look at the repository’s commit history and open issues.
Security reports
No reports found for this package in our research (last checked 2026-10-30).
Risks to watch for
- Skill injection: Malicious or poorly written skills added to the collection could execute unintended code when an AI agent runs them.
- Data leakage through agent outputs: Skills that process secrets, API keys, or sensitive files may accidentally expose them in AI-generated responses or logs.
- No built-in validation: The framework doesn't automatically check whether a skill is safe or has been tampered with before deployment.
- Service layer exposure: If skills are deployed as a service, weak authentication or improper network access controls could let unauthorized users run them.
- Prompt injection in skill design: Skills triggered by user input might be fooled by malicious prompts designed to bypass their intended behavior.
How to lower the risk
Least-privilege permissions
- Only grant skills access to the specific files, APIs, and accounts they need.
- Run agent processes with minimal OS-level permissions (non-root user, restricted directories).
Network isolation
- Keep skill services behind a firewall or authenticate all requests.
- Never expose skill services directly to the internet without strong authentication.
Monitoring
- Log what data each skill accesses and what it outputs.
- Review agent logs regularly for unexpected file access or API calls.
Red flags to watch for
- Skills that request credentials or API keys directly from prompts.
- Unusual file access patterns or network connections from deployed skills.
- Skills that echo back user input without filtering in agent responses.
Questions
How do I install Skills?
Skills does not publish a standalone npm or PyPI package. Clone or follow the setup instructions in its source repository at github.com/michaelshimeles/skills.
Are there security reports about Skills?
No reports found for this package in our research (last checked 2026-10-30). Skills currently holds Grade C (Moderate risk signals) based on public adoption and transparency signals. Key risk areas to watch include skill injection, data leakage through agent outputs, no built-in validation.
Is Skills actively maintained?
Skills has 1.4k GitHub stars. Our directory data was last refreshed on 2026-10-25. Check recent commit activity, release tags, and open issues at github.com/michaelshimeles/skills before pinning it in production.
What are alternatives to Skills?
Notable alternatives to Skills include LangChain tools and agents (Framework for building LLM tools with built-in review patterns), Semantic Kernel (Microsoft) (Structured skill composition for AI agents with validation), Native Claude API with tools (Anthropic's tool-calling feature without additional abstractions), Agent frameworks like AutoGPT or BabyAGI (Full agent orchestration (may have their own security gaps)). You can also compare Skills side by side with other skills in the Related tools section on this page.
README badge and team notes
Maintain or use Skills? Show its live grade in your GitHub README, docs site, or run a check from the terminal before you install.
GitHub README Markdown:
[](https://agentarmorhq.com/skills/michaelshimeles-skills)HTML embed:
<a href="https://agentarmorhq.com/skills/michaelshimeles-skills"><img src="https://agentarmorhq.com/api/badge/michaelshimeles-skills" alt="AgentArmorHQ Risk Grade for Skills" height="20" /></a>Terminal check:
curl -fsSL https://agentarmorhq.com/api/cli | node --input-type=module - check "michaelshimeles-skills"Share with your security lead
Need sign-off to use skills? Copy this note for Slack, a ticket or a pull request.
### AgentArmorHQ Security Review Memo: skills
- **Listing URL:** https://agentarmorhq.com/skills/michaelshimeles-skills
- **Type:** AI agent skill
- **Risk Grade:** Grade C (Moderate risk signals)
- **Outside Security Advisories on File:** 0
- **Public Adoption:** 1.4k GitHub stars
#### Permission & Blast-Radius Signals
- **Local files:** No local folder access noted
- **Command execution:** No shell command runner noted
- **Network reach:** Connects to outside service API
- **Credentials:** No API token required
- **Package source:** GitHub (michaelshimeles/skills)
#### Recommended Hardened Setup
- **Source Repo:** https://github.com/michaelshimeles/skills
- **Controls:** Pin an exact version in `.mcp.json`, scope access to a single workspace folder, and use a least-privilege read-only token where supported.
_Based on public adoption and transparency signals as of Oct 22, 2026. Our opinion, not a code audit or security guarantee._Key facts
- Type
- AI agent skill
- Publisher
- michaelshimeles
- Source code
- github.com/michaelshimeles/skills
- Est. context load
- ~78 tokens idle · ~450 tokens when active (on-demand skill)
- Popularity
- 1.4k GitHub stars
- Last updated
- 2026-10-25
- Last graded
- 2026-10-22
Community
Discussions for Skills
Ask questions or share configuration tips. Governed by our Community Rules & Terms.
No posts match this filter yet.
Be the first to ask a question or share a working tip.
Check before you install, in Claude
Your assistant answers Block, Review or No known issues for any package before it installs it.
Get this check inside Claude