Back to trending
How scoring works 
Self-hosted runtime control plane for AI agents. Observe or HITL approve or Block rogue tool calls before it executes: secret leaks, prompt injection, supply chain etc in a local dashboard. Agent agnostic (Claude, codex, langchain etc.)
Python401 stars52 forksApache-2.0
agent-control-planeagent-securityagentsai-agentsai-governanceai-safetyai-securityclaude-codecontrol-planecybersecuritydevsecopsllm-securitymcpowaspprompt-injectionsecrets-managementsecurityskill-scannersupply-chain-securitytool-calls
Rankings
PeriodRankScore
Daily
#29152.4Weekly
#138344.1Monthly
#184390.8Engagement
Total stars
Daily new stars
Embeds
Best rank #150