# ShenSeanChen/waku-agent

Waku Waku! Waku Agent is a local-first AI agent harness you actually own, including loop, memory, eval, all in code built to stay legible as it grows.

Repository: https://github.com/ShenSeanChen/waku-agent
Canonical: https://ross.abutalabs.com/products/waku-agent
Language: Python
License: MIT
License Family: permissive
Topics: agent-framework, ai-agent, ai-assistant, claude, evals, llm, llmops, local-first, openai, personal-assistant, python, sqlite
Last push: 2026-08-17T00:22:16+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 95, longevity 3
- inputs: {"age_days": 54, "days_push": 17, "days_rel": 33, "gap_med": 4, "n_releases_24m": 2}
- flags: young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1580, forks 292 (observed 2026-08-28T04:05:06.087620+00:00)

## What it is
Waku Agent is a local-first AI agent harness in Python that exposes the full agent stack — harness, loop, memory, and eval/LLM-Ops — as small, readable code. Memory lives in a single SQLite file with semantic, episodic, and procedural layers, and it ships with a terminal chat, a browser dashboard, and built-in deterministic plus LLM-as-judge evals.

## Use cases
- build a personal AI assistant that runs locally on my laptop
- learn how an agent loop and memory system work by reading real code
- run an LLM agent with persistent memory stored in sqlite
- evaluate my agent with deterministic tests and LLM-as-judge before releasing
- watch an agent's message flow in a local dashboard
- self-host a chatbot with a wake word and Telegram gateway

## When to choose
- you want a legible, hackable agent codebase instead of an opaque framework
- you want local-first ownership of your agent's memory as a plain SQLite file
- you want evals and a release gate built into your agent harness
- you're learning agent architecture (loop, memory, eval) from a working example

## When to avoid
- you need a production-grade, battle-tested multi-agent orchestration platform
- you need distributed or multi-user agent deployment at scale
- you want a no-code or fully managed agent builder
- you need non-Python runtimes or heavy vector-database integrations out of the box

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, rag, chatbot, testing, monitoring, database
- domain: artificial-intelligence, large-language-models, developer-tools, self-hosted
- platform: python, cli, cross-platform, self-hosted
- tags: local-first, agent-loop, agent-memory, llm-evals, llmops, sqlite-memory, personal-assistant, telegram-gateway, readable-code, ai-agents

## Member repositories
- ShenSeanChen/waku-agent (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:06.087620+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:56:36.223085+00:00, confidence not recorded.
  - readme: https://github.com/ShenSeanChen/waku-agent (fetched 2026-08-28T04:05:06.087620+00:00, sha 85c35dedc10b)
  - registry_pypi: https://pypi.org/pypi/waku-agent/json (fetched 2026-08-29T11:26:56.556263+00:00, sha 43515c4050a2)
- Data as of 2026-08-30T08:39:29.467469+00:00.