This is a production-grade long-term memory plugin for the OpenClaw AI agent platform, built as a TypeScript/ESM module. It provides a complete memory lifecycle: LLM-powered extraction from conversation transcripts (6-category classification with L0/L1/L2 layering), hybrid vector+BM25 retrieval with cross-encoder reranking, Weibull decay-based importance scoring, tier management (core/working/peripheral), and multi-scope isolation (global, agent, user, project). The plugin integrates deeply into OpenClaw via lifecycle hooks, agent tools, CLI commands, and a dreaming engine for periodic consolidation. It stores all data in LanceDB and supports multiple embedding providers, LLM backends, and OAuth flows. The codebase is large and complex, with extensive test coverage (100+ tests) and many edge-case handlers, but exhibits significant technical debt in a single massive index.ts file and multiple fallback/legacy migration paths. It is not a research project; it is a highly functional, actively maintained plugin with a clear engineering focus on correctness, resilience, and performance.
This repository provides a credential-free, deterministic test fixture for the OpenClaw plugin API surface, known as the 'kitchen sink' plugin. It exercises the full public plugin API by registering commands, tools, providers (image, speech, realtime transcription/voice, video, music, web search/fetch, embedding, memory, compaction), channels, hooks, gateway methods, services, and CLI backends. The plugin supports three personalities ('full', 'conformance', 'adversarial') to validate different plugin compatibility modes. It is designed as a compatibility fixture for OpenClaw's plugin-inspector and Crabpot tools, and is published to npm as @openclaw/kitchen-sink. The codebase is well-structured, with separate runtime modules per surface, comprehensive deterministic scenario routing, and a robust set of test scripts that validate runtime behavior, contract probes, and package metadata.
ClawBands is a security middleware for OpenClaw AI agents that provides runtime interception with human-in-the-loop approval. It hooks into OpenClaw's plugin system to intercept tool executions (file operations, shell commands, network requests) and enforces policy-based decisions before allowing potentially dangerous actions. The system operates in two modes: terminal mode with interactive prompts, and channel mode (WhatsApp/Telegram) using a special response tool. Every decision is logged to an immutable JSON Lines audit trail, and statistics are tracked. The architecture features clean separation of concerns with distinct modules for policy interception, decision arbitration, approval queuing, logging, and storage.
This repository is a curated community collection of 42 real-world use cases for OpenClaw (an AI agent framework), organized as markdown documentation. It is NOT a data pipeline, application, or research codebase itself — rather, it's an 'awesome list' style knowledge base documenting how practitioners use OpenClaw for tasks spanning social media automation, content pipelines, infrastructure management, personal productivity, research, and finance. Each use case file describes a workflow pattern, required skills/plugins, setup prompts, and architectural insights. The repo serves as a pattern catalog for agent-based automation. Classification as 'data_research' reflects the content's focus on data/engineering workflows (RAG, ETL, research pipelines, analytics), though the repo itself contains only documentation, not executable code or data.