# Vexa

## Docs

- [Agent API](https://docs.core.vexa.ai/api/agent.md): The control plane — dispatch a unit, stream a chat turn, manage routines, read the workspace.
- [Errors](https://docs.core.vexa.ai/api/errors.md): Status codes, error shape, and what each one means across the Meetings and Agent APIs.
- [Meetings API](https://docs.core.vexa.ai/api/meetings.md): Send a bot to a meeting, get the real-time transcript, manage the bot, retrieve per-speaker audio.
- [Architecture as Code (CALM)](https://docs.core.vexa.ai/architecture/architecture-as-code.md): The runtime architecture is a validated FINOS CALM model — nodes, data carriers, single-writer ownership, and a self-hostable meeting-intelligence pattern, enforced in CI.
- [The dispatch](https://docs.core.vexa.ai/architecture/dispatch.md): One agent run — the composition of the seven primitives.
- [Execution](https://docs.core.vexa.ai/architecture/execution.md): The agent runs in a runtime-spawned container, over a mounted workspace folder, resuming from a session file.
- [Governance](https://docs.core.vexa.ai/architecture/governance.md): Two axes — input trust × effect reversibility. Untrusted or irreversible → propose, approve, apply.
- [Identity & trust](https://docs.core.vexa.ai/architecture/identity-and-trust.md): Chain of custody — authenticity of the launcher, a signed token through every hop, verified at the boundaries.
- [Modules & Seams](https://docs.core.vexa.ai/architecture/modules.md): The single map of the codebase: the module tree, the sealed contract registry (owner → consumers), and the four eval levels.
- [Streaming](https://docs.core.vexa.ai/architecture/streaming.md): Redis Streams from the agent container, relayed to the terminal over websocket.
- [Authentication](https://docs.core.vexa.ai/authentication.md): Mint an API key, send it on every request, scope it, and rotate it.
- [Changelog & migration](https://docs.core.vexa.ai/changelog.md): Release notes and what changes between major versions.
- [Concepts](https://docs.core.vexa.ai/concepts.md): The primitives Vexa is built from — workspace, agent, container, identity, scheduler.
- [Configuration](https://docs.core.vexa.ai/configuration.md): Every environment variable the stack reads — what it does and its default.
- [Agents](https://docs.core.vexa.ai/core/agents.md): A sandboxed CLI agent that works a person's knowledge like a developer works a codebase.
- [Identity](https://docs.core.vexa.ai/core/identity.md): The one place that decides who is asking, what they own, and whether they are allowed — for every request in the system.
- [Meetings](https://docs.core.vexa.ai/core/meetings.md): A standalone service that captures, stores, and serves meeting data in real time — usable on its own or with agents.
- [Runtime](https://docs.core.vexa.ai/core/runtime.md): Isolated, ephemeral containers — the execution layer.
- [Deployment](https://docs.core.vexa.ai/deployment.md): Self-host Vexa with Docker Compose — air-gapped, with bring-your-own inference.
- [Chat with your workspace](https://docs.core.vexa.ai/how-to/chat-workspace.md): Ask an agent that has your whole knowledge base as context — and let it record what you decide.
- [Brief me every morning](https://docs.core.vexa.ai/how-to/daily-brief.md): An unattended agent that runs on a schedule and commits to your workspace.
- [Triage incoming email](https://docs.core.vexa.ai/how-to/email-triage.md): An event-triggered agent that turns new mail into proposed tasks — safely.
- [Run a live meeting copilot](https://docs.core.vexa.ai/how-to/live-copilot.md): Surface people, action items, and decisions as a call happens — approve them into the workspace.
- [How-to guides](https://docs.core.vexa.ai/how-to/overview.md): Task-oriented recipes, split by which plane they touch — meetings, agents, or both.
- [Report after every meeting](https://docs.core.vexa.ai/how-to/post-meeting-report.md): When a call ends, an agent writes notes, decisions, and follow-ups into your workspace.
- [Retrieve a meeting recording](https://docs.core.vexa.ai/how-to/recordings.md): List recordings and stream the audio — from your own object storage on a self-host.
- [Send a bot and get a transcript](https://docs.core.vexa.ai/how-to/send-a-bot.md): Capture a Google Meet, Zoom, or Teams call in real time.
- [Stream a transcript over WebSocket](https://docs.core.vexa.ai/how-to/stream-transcript.md): Live, per-segment transcript push instead of polling.
- [Browse, upload, and swap the workspace](https://docs.core.vexa.ai/how-to/workspace-files.md): Read the knowledge base over the API, add documents, or attach your own git repo.
- [Vexa](https://docs.core.vexa.ai/index.md): Sandboxed AI agents that grow corporate knowledge from your meetings and docs — open-source, self-hosted, air-gappable.
- [Quickstart](https://docs.core.vexa.ai/quickstart.md): From zero to your first transcript and your first agent — in about five minutes.
- [Approach](https://docs.core.vexa.ai/roadmap/approach.md): How we build — principles first, execution core before features, honest re-baselining.
- [Stages](https://docs.core.vexa.ai/roadmap/stages.md): The build, in dependency order. Each stage names what it delivers and the proof that closes it.
- [Status](https://docs.core.vexa.ai/roadmap/status.md): Where we are — honestly.
- [SDKs & clients](https://docs.core.vexa.ai/sdks.md): The canonical interface is the REST API; published libraries wrap it.
- [Troubleshooting](https://docs.core.vexa.ai/troubleshooting.md): The common failures, what causes them, and how to fix them.