Item detail

hashgraph-online/hol-guard

HOL Guard positions itself as an AI-agent security scanner for Codex, Claude Code, Cursor, Gemini, OpenCode, plugins, skills, MCP servers, and AI harnesses. The repo documents PyPI packages, plugin scanning, CLI workflows, and checks meant to catch risky agent/tool behavior before execution.

Score8.2
Popularity76.0
Riskconditional
TierGold
Score breakdown
Usefulness8.0
Novelty8.0
Momentum8.0
Maturity8.0
Open-source/build7.4
Evidence7.2
Workflow potential9.3
Setup ease6.4

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for teams experimenting with agent plugins and MCP servers: run it against a sandbox plugin/tool set, inspect the findings manually, and use it as one layer in a broader review process rather than as a magic safety guarantee.

Who should use it

AI agent usersMCP server evaluatorssecurity-conscious developer teamsplugin authors

Who should skip it

Skip if the source link, docs, or setup requirements do not match your workflow.

Risk explanation

GitHub metadata reports no standardized license; security tooling should be validated against source code and not treated as complete protection.

Evidence links

Closest alternatives / related signals

agent-securitymcpplugin-scannercodexclaude-codecursor