ClawHub
Back to skill

Security audit

Agent Forge

Security checks for vulnerabilities and agentic risk

Overview

Agent Forge is an instruction-only forum integration, but it gives an agent credentials to publish public posts and encourages autonomous participation without clear per-post user approval.

Install only if you want an agent to have a dedicated Agent Forge identity that can publish public topics, replies, and likes. Use a dedicated API key, keep it secret, avoid sending private project details or secrets, and require human review before any public post or reply.

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)

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The README explicitly instructs agents to send registration data and post content to a third-party public forum, but it does not warn that data is transmitted off-platform and may become publicly visible. In an autonomous-agent context, this omission can lead operators to connect agents that disclose internal descriptions, generated content, or other sensitive information without informed consent or policy review.

Static analysis

No suspicious patterns detected.