mattijsmoens/intentshield
Pre-execution intent verification for AI agents. Audits what your AI is about to do, not what it says. Zero dependencies, deterministic, hash-sealed.
This tool helps safeguard AI agents by auditing proposed actions (like running shell commands or writing files) before they execute. It takes an AI's intended action and its payload as input and determines if it's safe or dangerous, blocking harmful activities. Anyone deploying AI agents in sensitive environments, such as operations engineers, security professionals, or product managers, would find this useful for preventing malicious actions or data leaks.
Available on PyPI.
Use this if you need a robust, deterministic safety layer that prevents your AI agent from performing dangerous actions like deleting files, executing shell commands, or leaking sensitive information, even when content filters fail.
Not ideal if you need a built-in language model output parser or advanced hallucination detection, as these features have been removed to focus solely on action auditing.
Stars
17
Forks
3
Language
Python
License
—
Category
Last pushed
Mar 11, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/mattijsmoens/intentshield"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Featured in
Higher-rated alternatives
ucsandman/DashClaw
🛡️Decision infrastructure for AI agents. Intercept actions, enforce guard policies, require...
Dicklesworthstone/destructive_command_guard
The Destructive Command Guard (dcg) is for blocking dangerous git and shell commands from being...
microsoft/agent-governance-toolkit
AI Agent Governance Toolkit — Policy enforcement, zero-trust identity, execution sandboxing, and...
vstorm-co/pydantic-ai-shields
Guardrail capabilities for Pydantic AI — cost tracking, prompt injection detection, PII...
Pro-GenAI/Agent-Action-Guard
🛡️ Safe AI Agents through Action Classifier