ClawHub
Back to skill

Security audit

Evidence Gate

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed evidence-checking helper that may auto-activate for strong claims, but it does not execute code, persist state, or access sensitive data.

Install this if you want the agent to add structured evidence checks before strong conclusions or high-impact recommendations. Expect occasional extra gating or softened wording because implicit invocation is enabled.

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

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The trigger condition is intentionally broad and includes many common agent behaviors such as presenting conclusions, safety assertions, and recommendations. In practice this can cause the skill to self-invoke in routine analytical workflows, leading to prompt bloat, unnecessary gating, delayed responses, or recursive/meta-analytical behavior that disrupts normal operation. The skill's purpose is safety-oriented, so this is not malicious, but the breadth still creates a real workflow-integrity risk.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill enables implicit invocation, but the activation guidance in the metadata is broad and natural-language only, with no enforceable scope constraints in the file itself. That creates a risk the agent will invoke this gate in unintended contexts, influencing outputs about safety, diagnosis, rollback, or destructive actions without a tightly bounded trigger condition.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.