Discovered · unverified
agent-guard
**The safety check an AI agent runs *before* it acts.** Every agent that installs packages, runs shell commands, or commits/outputs text can do real damage — install malware, wipe a disk, or leak a credential. agent-guard is the cheap "look before you leap" gate an agent calls before the irreversible step. | tool | call it before… | |---|---| | check_package | pip/npm install — is it real, malware, a typosquat, or does it run code on install? | | check_command | running a shell command — destructive / RCE (rm -rf, curl\|bash, dd, force-push)? | | scan_secrets | committ
Check the source before connecting.
Vendor ownership, pricing, documentation, and execution are not verified for this discovery record. Appearing in the official MCP registry does not mean this server is vendor-owned or tested.
Relevance and review status.
Awaiting triage. Method: not yet classified.
Task signals: payment.
Setup and documentation.
No verified installation steps or callable endpoint are supplied by this discovery record. Follow the original repository or directory entry to inspect current setup instructions, permissions, and authentication.
Source receipts.
- smithery · Observed 2026-10-17
Discovery ID: 52e2e224bad21be80abd. Snapshot: 2026-10-17. Read this record as JSON · Full discovery dataset.
Submit primary documentation for review · Connect your agent to GTMMCPs