ClawHub
Back to skill

Security audit

Openclaw Agent Skill Evolution

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed framework for improving OpenClaw skills with testing and rollback safeguards, with no hidden installer or executable payload found.

Install only if you want an agent to help evaluate and improve OpenClaw skills. For shared, production, security-critical, credential-related, financial, or destructive skill changes, require explicit human review even though the skill permits limited low-risk auto-deploy.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The usage guidance says to use the skill 'when the conditions in the description are met,' but the trigger scope is not concretely defined in the README itself. For an agent skill that performs evolution, benchmarking, red-teaming, and iterative refinement, this ambiguity can cause over-broad activation and unintended invocation on tasks the user did not clearly authorize, increasing the chance of unsafe or unnecessary autonomous behavior.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
| Candidate menyebabkan CRITICAL REGRESSION | REJECT | Capability gain ≠ izin merusak fitur lama |
| Motif = HYPE/TRENDING/VIRAL tanpa evidence | BLOCK | No-Hype Rule |
| "Lebih panjang/complex" diklaim = "lebih pintar" | TOLAK klaim | No-Fake-Intelligence Rule |
| Low risk + reversible + verifiable + test passed | Boleh AUTO-DEPLOY | Otomatis-terbatas boundary aman |
| Perubahan kritis | PROPOSE → human approval → deploy | Agent tak ubah boundary sendiri |

## 7. REASONING POLICY
Confidence
90% confidence
Finding
The policy explicitly permits AUTO-DEPLOY when the agent judges a change to be low risk, reversible, verifiable, and tested. Even with guardrails, this authorizes autonomous modification of operational behavior based on the agent's own assessment, which can be wrong under adversarial input, incomplete context, or flawed tests. In a skill-evolution framework, that increases the chance of unauthorized or unsafe changes propagating without human review.

Static analysis

No suspicious patterns detected.