技能列表

Generischer Skill für Memory-Management, Hybrid Retrieval und Skill Evolution mit PluginEval Integration

End-to-end Meta-Harness evolution for Hoss (OpenClaw agent). Runs nightly at 3 AM via OpenClaw cron. Reads Hoss's current workspace configs (SOUL.md, IDENTITY.md, AGENTS.md, TOOLS.md, MEMORY.md), proposes harness modifications via a coding-agent proposer, evaluates against a benchmark, logs results to ~/hoss-evolution/, and posts a summary to the #research Discord channel. Triggered: (1) automatically via cron at 3 AM CDT, (2) when Tyler says "run harness evolution", "evolve Hoss", or "run meta-

>

|

|

Use the MeshMonitor REST API to inspect Meshtastic mesh state, nodes, channels, telemetry, messages, traceroutes, packets, solar data, and network-wide stats. Use when the user wants data from a MeshMonitor instance, wants a MeshMonitor-backed dashboard/report, wants help querying mesh history, or wants a reusable MeshMonitor API integration/skill created or updated.

Provide core mentoring without dependency. Clarify situations, reveal patterns, reframe choices, and return agency to humans seeking guidance.

|

完整记忆系统 - 四类记忆分类 + AutoDream自动整合 + MiniMax LLM驱动 + Feedback双向记录

Core memory resilience layer: WAL (Write-Ahead Log), Working Buffer, and three-layer memory integration. Prevents context loss during compaction and ensures critical state survives session restarts. Works with any OpenClaw agent.

Use when the user wants to review auto-memory, promote durable instructions into CLAUDE.md or local memory, and clean up duplicates or conflicts.

Configure OpenClaw's memory-lancedb plugin for semantic vector memory using a local LanceDB database. Use when: (1) setting up vector memory for the first time, (2) memory-lancedb fails with module not found errors, (3) migrating from flat-file MEMORY.md to vector-based recall, (4) configuring an embedding provider (Gemini, OpenAI-compatible). NOT for: general memory_store/memory_recall usage (just use the tools directly).