runtime-guardrails

Centralized agent control plane for governing runtime agent behavior at scale. Configurable, extensible, and production-ready.

Deterministic execution authorization for AI agents and automation systems

Stop Claude Code from doing irreversible damage. Policy-gated execution + receipts so you can ship agents without sweating production.

Enforces fine-grained permissions on agent tool calls and data access scopes

Enforces fine-grained permissions on agent tool calls and data access scopes

Python runtime checks for AI agents: budgets, loops, retries, and local traces. Zero runtime dependencies. MIT.

Production runtime guardrails for AI agents: budget caps, circuit breakers, and OpenTelemetry traces.

Trusted autonomy T&E runtime that links mission needs, hazards, scenarios, telemetry, evidence, verification reports, and hash-chained ledgers so AI/autonomous decisions can be reviewed instead of merely trusted.

Irreversible action classifier for autonomous AI agents. Zero-dependency, sub-5ms

Runtime guardrails for AI agents that enforce token budgets, loop limits, and tool rate limits locally.

AI agent security SDK for runtime guardrails: control tool calls, source-chain provenance, budget limits, plan preflight, prompt-injection isolation, and adjudication tokens. Apache-2.0 Python SDK.

AI Governance: The Foundation for Organized AI in Production — A Field Guide by Fabio Bastos. Published under CC BY 4.0.

🌿🐡"51 deterministic checks across 13 areas to catch bugs and harden your code."

Stage0 authorization gate for LangGraph agents with Human-in-the-Loop support - demonstrates ALLOW/DENY/DEFER routing, interrupt/resume workflows, and state persistence

Enforce token budgets, loop limits, and tool rate limits in AI agent runtimes to control execution and prevent resource overuse.