ClawHub
Back to skill

Security audit

agent-instruction-slimming

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill benign with medium confidence; scanners found no AIG alerts and treated the single low-probability Skillspector flag as a caution, though one reference was unresolved.

The verdict is benign, but confidence is medium and the scan is partial: a local path-like reference in SKILL.md could not be resolved. Consider that gap and the note about approval wording in the examples when deciding whether to install.

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 (2)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% 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 · examples/original.md (reported line 5)May include surrounding context.

md
## Safety

- Never send data to an external endpoint without approval.
- Do not store credentials in files or command lines.
- Destructive commands must be explained before execution.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% 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 · examples/slim.md (reported line 5)May include surrounding context.

md
## Safety

- Never send data to an external endpoint without approval.
- Do not store credentials in files or command lines.
- Destructive commands must be explained before execution.