ClawHub
Back to skill

Security audit

Hubo

Security checks for vulnerabilities and agentic risk

Overview

Hubo is a disclosed coding-review orchestration skill, with the main caution that it mirrors detailed work and review reports into the current conversation.

Use this skill when you are comfortable with a work agent editing task-relevant files and a review agent inspecting the same code. Avoid using it on highly sensitive repositories unless you are comfortable with detailed diffs, commands, evidence, and review commentary appearing in the conversation transcript.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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)

Ssd 3

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

These instructions require the coordinator to mirror full role reports into the user-visible conversation and preserve detailed findings, commands, evidence, and decisions, while only explicitly redacting credentials, secrets, or system-protected data. That creates a real risk of over-disclosing sensitive but non-secret user or repository data contained in reports, diffs, logs, test output, or reviewer commentary, especially because the skill orchestrates broad inspection of a codebase and asks for comprehensive traceability.

Content

No source excerpt is available for this finding.