pipecat-unison
llms.txt0 starsChanged 5 months agoMITView on GitHub
Drop-in persistent memory for Pipecat voice AI agents, powered by the Unison brain — a hosted, path-addressable knowledge base with hybrid semantic + keyword search. Agents: start with AGENTS.md — it covers install, auth with a usk_ key, programmatic usage, and how to contribute to this repo. The Pipecat processor (backend/unison_pipecat.py) intercepts LLMContextFrame objects in your pipeline, retrieves relevant memories from the Unison brain, injects them into the system prompt, then saves a session note after each turn so knowledge…
Files
# pipecat-unison > Drop-in persistent memory for [Pipecat](https://pipecat.ai) voice AI agents, powered > by the [Unison brain](https://unisonlabs.ai) — a hosted, path-addressable knowledge > base with hybrid semantic + keyword search. **Agents: start with [AGENTS.md](AGENTS.md)** — it covers install, auth with a `usk_` key, programmatic usage, and how to contribute to this repo. The Pipecat processor (`backend/unison_pipecat.py`) intercepts `LLMContextFrame` objects in your pipeline, retrieves relevant memories from the Unison brain, injects them into the system prompt, then saves a session note after each turn so knowledge accumulates across conversations. Auth is a `usk_...` machine key via `UNISON_TOKEN` env var or `token=` constructor argument. The backend is the security boundary; the client enforces nothing. ## Docs - [README](README.md): quickstart, environment variables, pipeline usage, deployment - [AGENTS.md](AGENTS.md): how an agent installs, authenticates, and uses this client - [CONTRIBUTING.md](CONTRIBUTING.md): development workflow, conventions, PR checklist ## Usage - [backend/unison_pipecat.py](backend/unison_pipecat.py): `UnisonPipecatService` — the Pipecat frame processor - [backend/server.py](backend/server.py): FastAPI + Pipecat voice server with full pipeline - [src/app.tsx](src/app.tsx): React frontend — voice UI + live Unison brain memory panel ## Examples - [README — Pipecat processor](README.md#pipecat-processor): minimal pipeline with `UnisonPipecatService` - [README — Getting a token](README.md#getting-a-unison_token): provision a `usk_` key in one `curl` call
In this repositoryOne other file it gives its agents.
- AGENTS.mdAGENTS.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.