ClawHub
Back to skill

Security audit

golang-cli

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent Go CLI development guidance skill with scoped examples and no evidence of hidden persistence, exfiltration, or deceptive behavior.

Before installing, expect the skill to guide an agent in editing Go CLI code and running normal Go, lint, and git commands. Review any proposed git command that would publish or rewrite repository history, but the package itself does not show hidden or automatic unsafe behavior.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • 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)

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · evals/evals.json (reported line 6)May include surrounding context.

json
"id": 1,
    "name": "minimal-main-and-execute",
    "description": "Tests that main.go is minimal and only calls Execute(), with os.Exit handled in main not inside commands",
    "prompt": "Create the entry point for a Go CLI application called 'deploy' using Cobra. The app should have a root command and a 'push' subcommand. Write main.go and root.go.",
    "trap": "Model puts configuration logic, flag parsing, or complex setup directly in main.go instead of keeping it minimal. May also call os.Exit inside RunE instead of returning errors.",
    "assertions": [
      {

Vague Triggers

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

This JSON manifest contains many prompt/description entries phrased as broad natural-language requests such as 'Write the configuration setup code' and 'What's the best approach?' without any explicit trigger constraints, exclusions, or invocation boundaries. In a manifest-scoped skill context, such generic language can overlap with common everyday requests and may contribute to unintended activation if these strings are used as matching examples.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.