ClawHub
Back to skill

Security audit

Autonomous Loops

Security checks for vulnerabilities and agentic risk

Overview

This Markdown-only autonomous coding workflow skill is not malicious, but its broad triggers and ungated commit steps warrant human review before installation.

Install only if you want agents to run broad autonomous coding loops. Require the agent to show the diff and ask before deleting tests, changing broad working-tree content, or creating commits.

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)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger section lists broad conditions such as iterative coding loops, redundant LLM output, and multi-step coding workflows without defining specific invocation phrases, exclusions, or negative examples. These conditions overlap with many ordinary software tasks, so the skill could be selected in unintended situations.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill explicitly instructs the agent to commit changes without requiring an approval gate or warning about persistent repository modification. In an autonomous loop, this can cause unintended state changes, premature commits of unsafe code, or persistence of sensitive/debug artifacts, especially when the workflow is designed to iterate automatically.

Content

No source excerpt is available for this finding.