ClawHub
Back to skill

Security audit

Subvisory

Security checks for vulnerabilities and agentic risk

Overview

The Subvisory controls are sensible, but setup can execute downloaded code and introduce broader agent changes without clear user approval.

Review and install the CLI yourself before use, and decide whether its automatic agent-skill synchronization is wanted. Trusting OOMOL to handle the connected account is part of this integration. Keep the required confirmations for changes and deletions. The downloaded installer payload was not independently inspected.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The supplied code performs no visible Subvisory operation. Instead, it executes a remote installer from an Oomol domain, with no demonstrated connection to supporting Subvisory access. The downloaded script's further behavior cannot be determined from this chunk.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
98% confidence
Finding

Piping a mutable remote script directly into bash executes its contents with the invoking user's privileges without independent integrity verification or review. Compromise of the installation endpoint or distribution infrastructure could enable credential theft or local file modification; HTTPS and the legitimate dependency context do not eliminate this supply-chain risk.

Content

Scanner excerpt · SKILL.md (reported line 69)May include surrounding context.

Fallback setup executes downloaded installation scripts without warning about remote code execution or system changes.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The fallback directs the agent to install software using downloaded scripts without requiring informed user approval for local code execution and system changes. Triggering setup only when the CLI is missing reduces exposure, but the Subvisory write-confirmation rules do not clearly cover this local installation.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 90)May include surrounding context.

md
- **`OOMOL_INSUFFICIENT_CREDIT`** — OOMOL billing stop. Recharge at `https://console.oomol.com/billing/token-recharge` before retrying.

- **HTTP 402 without `OOMOL_INSUFFICIENT_CREDIT`** — the third-party account itself has no active plan or credit. Report the provider error message as is; do not ask the user to recharge OOMOL.

## Resources