ClawHub
Back to skill

Security audit

handle-big-tasks

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently automates multi-phase coding work and clearly discloses that its optional loop scripts can keep Copilot running until the plan is complete.

Installers should understand that running the loop scripts can let Copilot continue work over many phases and, if you pass write or shell permissions through LOOP_COPILOT_ARGS, it may modify the workspace without interactive approval during each run. Use a reviewed plan file, keep permissions narrow, and avoid broad flags like allow-all-tools except in trusted repositories.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The script repeatedly invokes the Copilot CLI with arbitrary extra flags from LOOP_COPILOT_ARGS and explicitly instructs the agent to continue autonomously across multiple runs. In the context of a task-execution skill, this can cause file writes or tool execution without an explicit interactive confirmation or warning in the script, increasing the chance of unintended or unsafe changes if the plan file or model output is adversarial or mistaken.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.