Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Clawboard is a self-hosted logging and retrieval system for OpenClaw AI gateway traffic. It runs as a Docker stack with a Next.js web UI, FastAPI backend, Qdrant vector index, and a background classifier, capturing conversations through the clawboard-logger plugin. Retrieval combines dense vectors, BM25, and lexical matching with reciprocal rank fusion.
Keeps all conversation logs on your own infrastructure with no third-party data sharing, while still offering vector search, reranking, and a web UI.
When to use it
- Searching past OpenClaw conversations by topic or keyword
- Monitoring AI gateway response times across sessions
- Reviewing a full history of prompts and completions
- Backing up curated OpenClaw memories to a private GitHub repo
- Running a self-hosted conversation index without cloud exposure
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Install and configure Clawboard with my OpenClaw setup
- 1Run the bootstrap script to clone the repo and detect the OpenClaw workspace layout
- 2Generate CLAWBOARD_TOKEN and write environment variables to .env
- 3Start Docker services: web, api, classifier, and qdrant
- 4Install and enable the clawboard-logger plugin in OpenClaw
- 5Restart the OpenClaw gateway and validate API health and search endpoints
Clawboard running at localhost:3010 with logging active; search returning results from OpenClaw conversations