ClawHub
Back to skill

Security audit

abe-foundry

Security checks across malware telemetry and agentic risk

Overview

This skill is openly a self-modifying OpenClaw capability builder, but it asks for broad authority that should be reviewed before use.

Install only if you intentionally want a high-privilege, self-modifying OpenClaw development extension. Review the external npm package and repository first, use a sandbox or separate OpenClaw profile, keep auto-learning and marketplace publishing disabled unless needed, and require manual review before generated tools, hooks, skills, or self-updates are enabled.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

External Transmission

Medium
Category
Data Exfiltration
Content
API_BASE = "https://api.heybossai.com/v1"

def pilot(body: dict) -> dict:
    r = requests.post(
        f"{API_BASE}/pilot",
        headers={"Authorization": f"Bearer {SKILLBOSS_API_KEY}", "Content-Type": "application/json"},
        json=body,
Confidence
95% confidence
Finding
The skill explicitly sends user-supplied prompts, research queries, scraped content, or generated code requests to an external third-party API endpoint using an API key. In the context of a self-writing, self-modifying agent, this creates a real data exfiltration and privacy boundary issue because potentially sensitive workspace, prompt, or internal failure data may be transmitted off-host to an external service.

Self-Modification

High
Category
Rogue Agent
Content
1. **Research** — Fetch and understand OpenClaw documentation on demand
2. **Write Extensions** — Generate new tools and hooks for OpenClaw
3. **Write Skills** — Create ClawHub-compatible skill packages
4. **Self-Modify** — Add new capabilities to itself
5. **Learn** — Record patterns from failures and successes
Confidence
96% confidence
Finding
The skill advertises that it can write new extensions and skills, meaning it can generate executable capabilities that may later be installed or trusted by the agent ecosystem. In a meta-extension that learns from failures and external sources, this materially increases the risk of prompt injection, unsafe code generation, and supply-chain propagation of malicious or insecure artifacts.

Self-Modification

High
Category
Rogue Agent
Content
1. **Research** — Fetch and understand OpenClaw documentation on demand
2. **Write Extensions** — Generate new tools and hooks for OpenClaw
3. **Write Skills** — Create ClawHub-compatible skill packages
4. **Self-Modify** — Add new capabilities to itself
5. **Learn** — Record patterns from failures and successes

All AI inference (LLM, web search, scraping) is routed through **SkillBoss API Hub** using `SKILLBOSS_API_KEY`.
Confidence
98% confidence
Finding
Self-modification is especially dangerous here because the skill can add new capabilities to itself while also consuming external search, scraping, and learned experience. That combination can let untrusted external content influence future tool behavior, expanding privileges or persistence over time and creating a pathway for autonomous compromise or rapid spread of insecure functionality.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.