ClawHub
Back to skill

Security audit

Generic Quality Gateways for Unattended Agent Development

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed quality-gateway reporting template that reads project evidence and writes bounded reports/configuration files inside the repository.

Before installing, expect the skill to inspect project evidence and create or update quality-gateway files in your repository. Run it on a clean branch or review generated docs/quality, .defs, and .tmp/quality-gates changes before committing.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly directs the agent to write reports, evidence, and temporary files into repository paths, but it does not require an explicit user confirmation before modifying repository contents. This can lead to unintended repository changes, noisy diffs, or accidental inclusion of generated artifacts in commits, especially when run in automated or semi-automated environments.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

This JSON file is a manifest-style template, so SQP-1 applies. The structure enumerates gates and checks, but it does not specify when this template should be used, what contexts it applies to beyond a generic project profile, or any negative examples/exclusions; that can make activation or selection criteria overly broad in downstream skill workflows.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.