> ## Documentation Index > Fetch the complete documentation index at: https://docs.plyra.dev/llms.txt > Use this file to discover all available pages before exploring further. # Overview > Policy enforcement middleware for AI agent tool calls AI agents execute real actions — deleting records, calling APIs, sending emails. There is no standard safety layer between the LLM's decision and execution. plyra-guard is that layer. It intercepts every tool call your agent makes, evaluates it against your policy, and blocks, logs, or escalates — before anything irreversible happens. plyra-guard runs **in-process** — no sidecar, no network hop. Every evaluation completes in under 2ms.
Plyra Guard Architecture
## How it works Every tool call passes through plyra-guard before execution: 1. Your agent decides to call a tool 2. plyra-guard intercepts the call in-process 3. The call is evaluated against your policy — sub-2ms, no network hop 4. Verdict: **ALLOW**, **BLOCK**, **ESCALATE**, **DEFER**, or **WARN** 5. The decision is written to the audit log Evaluation runs in-process. No network hop. Sub-2ms per call. Works with LangGraph, AutoGen, CrewAI, LangChain, OpenAI, Anthropic, and plain Python. Rules live in your repo, reviewed in PRs, tested in CI. YAML or Python — your choice. Every decision logged — allowed and blocked. Ships to OTEL, Datadog, or your own sink. ## What's available today Install with `pip install plyra-guard`. Stable beta API. Python 3.10+. No telemetry. No usage tracking. Run it on-prem, in a container, or embedded in your agent loop. Native support for LangGraph, LangChain, AutoGen, CrewAI, OpenAI, Anthropic, and plain Python. Protect your first tool call in 60 seconds. Write rules that match your threat model.