ClawHub
Back to skill

Security audit

Agentic Coding

Security checks for vulnerabilities and agentic risk

Overview

This skill is a transparent coding-workflow guide that stores optional local notes and does not contain hidden execution, network, credential, or destructive behavior.

Before installing, decide whether you want the skill to activate only on request or during higher-risk coding tasks. If enabled, expect it to keep local workflow notes under ~/agentic-coding/; review or delete that folder if you do not want persistent coding preferences retained.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 125)May include surrounding context.

md
- Trigger undeclared network requests
- Access files outside its own memory path
- Write to global or platform memory stores
- Auto-approve risky code without explicit evidence

## Scope

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The markdown asks to clarify activation 'within the first exchanges' using broad conditions such as 'when the user asks for code changes with quality gates' and 'proactively when risk is high,' but it does not define concrete trigger phrases or clear boundaries. This can cause unintended invocation because the activation scope is described conceptually rather than with specific, narrow triggers or exclusion examples.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.