openclaw-setup

Curated, security-first OpenClaw skills (Markdown-based). Security audit skills - detect prompt injection, supply chain attacks, credential leaks. Works with Codex CLI, Claude Code, any LLM.

DSMC — Dual-State Multiversal Cognition: Minimal zero-dependency governance layer for OpenClaw agents. Maintains external state + classifies every message (DIRECTIVE/REVISION/EXAMPLE/etc.) to prevent context drift, revision failures, and example contamination in long-running sessions.

OpenClaw skill for scraping ChatGPT, Gemini, Perplexity and Grok responses. Built for AI agents and GEO monitoring.

Build a persistent, searchable memory system for OpenClaw that stores cross-session context to improve AI assistant recall and continuity.