HN Showcase
Back to browse
GitHub Repository

200 AI agent skills, hardened with targeted behavioral guardrails. Free drop-in replacements.

23 starsJavaScript

OpenClaw skills degrade agent safety

by shadab_nazar·Feb 26, 2026·1 point·2 comments
Visit ProjectView on HN

AI Analysis

●●●BangerBig BrainWizardryZero to One

Behavioral safety testing reveals 45 regressions static analysis misses—guardrails provided.

Strengths
  • •Discovers a real gap: static scanners (Snyk, semgrep) miss behavioral regressions from skills.
  • •Cross-model replication (186 test categories, 4,870 generations on Claude Opus) is rigorous methodology.
  • •Provides actual fixed code, not just warnings—82% regression fix rate with measurable improvements.
Weaknesses
  • •Tailored to OpenClaw ecosystem, limiting adoption to that tool's users.
  • •No automated testing harness provided—teams must manually integrate behavioral pipeline.
Category
Security
Target Audience

AI agent builders using OpenClaw, teams deploying agentic systems with untrusted skills

Similar Projects

AI/ML●●Solid

OpenClaw-superpowers – Self-modifying skill library for OpenClaw agents

Self-modifying skills let agents persist new behaviors without restarts or config edits.

Ship ItBig Brain
Arkid
803mo ago
Security●●Solid

SecureClaw – Open-Source Security Layer for OpenClaw Agents

The two-layer approach — a code plugin for gates/hardening plus a tiny ~1,230-token LLM skill for behavioral rules — is smart and practical. I appreciate that detection runs in bash (no token bloat) and that they mapped concrete checks to OWASP ASI and MITRE frameworks; the tradeoff is obvious: this is highly valuable if you run OpenClaw, but mostly irrelevant outside that ecosystem.

Niche GemBig Brain
alex_polyakov
213mo ago