SigmaShake
AI Agent Guardrails for Claude Code, Cursor, Gemini CLI, and Copilot. Deterministic policy before agents execute (ALLOW, DENY, ASK, FORCE). Policy at the tool-call boundary. Honest-agent / integration-dependent. Not a sandbox.
What it is
The problem it solves
Observed features
Evidence
Verified:
- sigmashake.comcrawl-v1 · 27deec84761f
Was this result useful?