Free Vetted AI SkillsTested. Trusted. Ready to Deploy.
Every skill is security-scanned and tested before listing. Download, install, and ship faster.
Browse Skills
Open source skills for OpenClaw agents — ready to install
Agent Sidecar
Observe-and-report diagnostics for any agent. Auto-discovers identity, flags critical issues and drift, produces copy-paste fix prompts. Two-phase: preflight builds context, diagnostics check memory health, self-healing, behavioral drift, integrations. Never auto-fixes.
AI Search Evals
Lightweight framework for evaluating AI search providers and RAG pipelines. Dual-route measurement, LLM-as-Judge, golden set builder.
Parallel Exec
Auto-detect whether to parallelize tasks or run single-threaded. 3-agent sweet spot, decomposability thresholds, merge strategies.
RLM Reader
Recursive Language Model pattern for processing arbitrarily long documents. Treats input as an external environment with search, slice, and parallel reading.
Cross Verify
Run any task through multiple AI models and cross-check results for errors, hallucinations, and blind spots. 2-model pass with judge synthesis, confidence calibration, and cost-benefit guidance.
Anthropic Financial Services
11 hardened autonomous agents for Wall Street — pitch decks, KYC screening, GL reconciliation. Audited by AgentVet with 4 critical patches applied.