ClawHub
Back to skill

Security audit

Skill Publish

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed ClawHub publishing helper that reads a user-selected skill folder and only publishes after explicit user confirmation.

Install this only if you intend to let an assistant inspect a local skill folder and help prepare ClawHub CLI commands. Review the generated publish command and release metadata carefully, and do not approve publishing unless the slug, version, changelog, tags, and target path are correct.

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
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (6)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 24)May include surrounding context.

md
- A target skill folder with a valid `SKILL.md`.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 70)May include surrounding context.

md
- A target skill folder with a valid `SKILL.md`.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 91)May include surrounding context.

md
- A target skill folder with a valid `SKILL.md`.

Self-Modification

High
Category
Rogue Agent
Confidence
85% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · eval.yaml (reported line 22)May include surrounding context.

yaml
- id: negative-trigger-seo-only
    prompt: |
      Rewrite this skill description so it ranks better on ClawHub search results, but do not prepare any publish command.
    expect:
      should_trigger: false

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The skill explicitly instructs the agent to read local files and execute shell commands (python3 ...check_publish_ready.py, clawhub publish, clawhub whoami, clawhub search) but does not declare any tool scope such as permissions or allowed-tools. This creates a permission-transparency gap: an agent or user may invoke a skill that can access the filesystem and run commands without the skill metadata clearly constraining or disclosing those capabilities, increasing the risk of unintended command execution or exposure of local data during publishing workflows.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/check_publish_ready.py (reported line 230)May include surrounding context.

python
findings.append({"severity": "medium", "message": "clawhub CLI is not installed or not on PATH."})
    else:
        try:
            whoami = subprocess.run(
                ["clawhub", "whoami"],
                capture_output=True,
                text=True,

Static analysis

No suspicious patterns detected.