# Apie

> Runtime visibility and guardrails for AI agents.

## Docs

- [Python sync and async](https://apie.mintlify.app/appendix/python-sync-and-async.md): When to use Apie vs AsyncApie, contextvar behavior, and async helper conventions.
- [Action and resource metadata](https://apie.mintlify.app/boundaries/action-and-resource-metadata.md): Model what your agent does (action) and what it touches (resource) so guardrails can evaluate risk.
- [Boundary drift](https://apie.mintlify.app/boundaries/boundary-drift.md): Detect when your agent uses undeclared tools or touches unknown resources in production.
- [Declare capabilities](https://apie.mintlify.app/boundaries/declare-capabilities.md): Tell Apie what tools, actions, and resources your agent is allowed to use.
- [How Apie works](https://apie.mintlify.app/concepts/how-apie-works.md): Understand the SDK lifecycle — register, enqueue, batch, and flush telemetry to Apie.
- [Choose how to instrument](https://apie.mintlify.app/getting-started/choose-how-to-instrument.md): Pick the right integration tier for your agent runtime — from zero-code MCP proxy to explicit per-tool metadata.
- [Connect your first agent](https://apie.mintlify.app/getting-started/connect-your-first-agent.md): Install the SDK, register your agent with Apie, and send your first telemetry event.
- [Enable guardrail templates](https://apie.mintlify.app/guardrails/enable-guardrail-templates.md): Turn on starter guardrail packs from the Apie dashboard via the CLI.
- [Enforce guardrails](https://apie.mintlify.app/guardrails/enforce-guardrails.md): Switch to Enforce mode to block risky actions and require human approval before execution.
- [Human approval](https://apie.mintlify.app/guardrails/human-approval.md): Pause agent execution when a guardrail requires human approval in the Apie dashboard.
- [Monitor mode](https://apie.mintlify.app/guardrails/monitor-mode.md): Observe guardrail evaluations without blocking your agent — the safe first step before enforcement.
- [Introduction](https://apie.mintlify.app/index.md): Runtime visibility and guardrails for AI agents — see what your agents do, constrain what they can do, and ship with confidence.
- [CrewAI, AutoGen, and LlamaIndex](https://apie.mintlify.app/integrations/crewai-autogen-llamaindex.md): Instrument CrewAI tasks, AutoGen steps, and LlamaIndex queries with Apie wrappers.
- [Integrations](https://apie.mintlify.app/integrations/index.md): Choose the right Apie integration helper for your agent framework or platform.
- [LangChain and LangGraph](https://apie.mintlify.app/integrations/langchain.md): Instrument LangChain agents and LangGraph nodes with Apie callback handlers.
- [LLM tool calls](https://apie.mintlify.app/integrations/llm-tool-calls.md): Wrap OpenAI and Anthropic native tool calls with Apie guardrails and telemetry.
- [OpenAI Agents](https://apie.mintlify.app/integrations/openai-agents.md): Instrument OpenAI Agents SDK runs with Apie lifecycle hooks.
- [Platform connectors](https://apie.mintlify.app/integrations/platform-connectors.md): Instrument GitHub, GitLab, issue trackers, incident tools, and observability platforms with canonical action metadata.
- [Vercel AI SDK](https://apie.mintlify.app/integrations/vercel-ai.md): Track LLM generations from the Vercel AI SDK with Apie telemetry.
- [Instrumented MCP client](https://apie.mintlify.app/mcp/instrumented-client.md): Wrap an in-process MCP client so tool calls are guarded and traced inside a run.
- [MCP proxy](https://apie.mintlify.app/mcp/proxy.md): Instrument MCP tool calls with zero code changes — point your MCP host at the Apie proxy.
- [Custom telemetry](https://apie.mintlify.app/observe/custom-telemetry.md): Emit workflow steps, LLM calls, MCP calls, handoffs, and errors beyond basic tool instrumentation.
- [Instrument tool calls](https://apie.mintlify.app/observe/instrument-tool-calls.md): Model tool, action, and resource metadata so Apie can trace and guard what your agent does.
- [Multi-agent pipelines](https://apie.mintlify.app/observe/multi-agent-pipelines.md): Model orchestrator and worker agents with sessions, child runs, and handoff events.
- [Runtime intelligence](https://apie.mintlify.app/observe/runtime-intelligence.md): Turn session telemetry into a readable story, action items, and guardrail recommendations.
- [Trace runs and sessions](https://apie.mintlify.app/observe/trace-runs-and-sessions.md): Wrap agent work in runs and sessions so every invocation appears in the Apie dashboard.
- [Boundary reports](https://apie.mintlify.app/production/boundary-reports.md): Generate agent boundary and compliance reports over a time window.
- [Diagnose your setup](https://apie.mintlify.app/production/diagnose-your-setup.md): Use doctor, send-test-event, and queue diagnostics to verify your Apie connection.
- [Redact secrets](https://apie.mintlify.app/production/redact-secrets.md): Strip API keys, tokens, and sensitive fields from telemetry before it leaves your process.
- [Reliable telemetry](https://apie.mintlify.app/production/reliable-telemetry.md): Tune the event queue, persist to disk, and shut down gracefully so no telemetry is lost.
- [Guardrail packs](https://apie.mintlify.app/recipes/guardrail-packs.md): Walkthrough — exercise five risky action scenarios in monitor mode to preview guardrail decisions.
- [Incident remediation](https://apie.mintlify.app/recipes/incident-remediation.md): Walkthrough — baseline monitor-mode agent investigating production incidents.
- [MCP enforcement](https://apie.mintlify.app/recipes/mcp-enforcement.md): Walkthrough — Enforce mode and approval routing through the MCP proxy.
- [Multi-agent handoff](https://apie.mintlify.app/recipes/multi-agent-handoff.md): Walkthrough — orchestrator triages a ticket and delegates response drafting to a worker agent.
- [Production release gate](https://apie.mintlify.app/recipes/production-release-gate.md): Walkthrough — pipeline session with LLM risk summary and high-risk CI/CD MCP trigger.
- [Runtime intelligence walkthrough](https://apie.mintlify.app/recipes/runtime-intelligence-walkthrough.md): Walkthrough - see the dashboard story generated from a production release session.
- [CLI reference](https://apie.mintlify.app/reference/cli.md): All Apie CLI commands and flags for TypeScript and Python.
- [Configuration reference](https://apie.mintlify.app/reference/configuration.md): Every ApieConfig field, environment variable, and config file format.
- [Errors reference](https://apie.mintlify.app/reference/errors.md): Apie error types, guardrail failure modes, and MCP JSON-RPC error codes.
- [Events reference](https://apie.mintlify.app/reference/events.md): Catalog of telemetry event types emitted by the Apie SDK.