ClawHub
Back to skill

Security audit

Validate Agent

Security checks for vulnerabilities and agentic risk

Overview

This skill is a quality-check checklist that may run often, but it does not request dangerous access or hide unrelated behavior.

Install this if you want an agent to spend extra attention on verification before finishing work. Expect it to add validation overhead and possibly create a simple promises.md tracking file for cross-session commitments; review or narrow the trigger language if you only want it used for higher-risk tasks.

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)

Vague Triggers

High
Confidence
96% confidence
Finding
The skill defines activation on very common workflow milestones such as task completion, post-fix, and before answering, which can cause it to run in a large fraction of interactions. Overly broad triggers increase the chance of unintended invocation, token/latency overhead, and workflow interference, especially because the skill influences validation and response behavior across many tasks.

Vague Triggers

High
Confidence
95% confidence
Finding
The trigger section repeats broad activation rules without exclusions or gating logic, making invocation ambiguous and potentially automatic in most sessions. This can create excessive validation loops, reduced usability, and unintended behavioral control over the agent's normal workflow.

Static analysis

No suspicious patterns detected.