ClawHub
Back to skill

Security audit

Executing Plans

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only workflow skill for following and verifying plans, with no code, persistence, credentials, or hidden execution behavior.

This skill is appropriate for users who want an agent to follow a written plan carefully. As with any planning workflow, review the plan itself before execution, especially if the plan includes destructive commands, account changes, or other high-impact actions.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 57)May include surrounding context.

md
## Anti-Patterns

- [ ] Do not improvise around a broken plan — amend it visibly or stop; silent drift is unaccountable work
- [ ] Do not skip verifications when steps "obviously worked" — the mysterious step-6 failure was born at step 3
- [ ] Do not push through a stop condition on momentum — it was written calm precisely because you wouldn't be
- [ ] Do not declare done without running the done-test — feeling-finished and being-finished diverge exactly when it matters
- [ ] Do not end a session without the state note — re-derivation is the tax on every resumed task

Static analysis

No suspicious patterns detected.