← Browse
mcp-security-guard
petrovicistefan ★ 0
mcp-security-guard audits MCP servers, plugins, skills, and commands for security threats mapped to the OWASP MCP Top 10, including tool poisoning, rug pulls, injected instructions, and secrets—all running locally with optional paid threat feeds and team policy. It hooks into tool use events to confirm credentials before execution and warn of injected instructions or hidden characters in responses.
Install
> /plugin marketplace add petrovicistefan/mcp-security-guard/tree/HEAD/plugin
> /plugin install mcp-security-guard
Source: https://github.com/petrovicistefan/mcp-security-guard/tree/HEAD/plugin
What it's made of
1 command · 1 skill · 1 MCP server · 3 hooks
- Commands
- 1
- Agents
- 0
- Skills
- 1
- MCP servers
- 1
- Hooks
- 3 · PostToolUse, PreToolUse, SessionStart
What it needs & plugs into
- API keys
MCP_SECURITY_GUARD_FEED_KEY- Paid services
- mcp-security-guard Team Plan
- External tools
node- Talks to
- nothing external detected
Analyzed . Facts extracted from the plugin's files. Prose generated by claude-haiku-4-5-20251001.
Is this your plugin?
Claim it to keep the card accurate and enter the weekly contest. Requires signing in as the GitHub owner (petrovicistefan).