ClawHub
Back to skill

Security audit

Portrait Generator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward external portrait-generation wrapper, with privacy considerations because portrait prompts and facial attributes are sent to ComfyDeploy.

Install only if you are comfortable sending portrait descriptions and structured facial attributes to ComfyDeploy. Avoid using it for real people or sensitive personal traits unless you have consent and understand the external service's privacy, retention, and access controls.

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 (3)

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill sends user-provided portrait prompts and highly sensitive facial attributes such as sex, ethnicity, age-related details, scars, deformities, and other biometric-like descriptors to a third-party ComfyDeploy API, but it does not clearly warn users that this data leaves the local system. This is a real privacy and transparency issue because users may unknowingly disclose sensitive personal or demographic information to an external service.

External Transmission

Medium
Category
Data Exfiltration
Content
## Endpoint

```
POST https://api.comfydeploy.com/api/run/deployment/queue
```

## Headers
Confidence
83% confidence
Finding
This endpoint definition confirms that the skill transmits user input to an external third-party service. In context, external transmission is expected for an image-generation skill, but it still creates a genuine security/privacy risk because the transmitted content may include sensitive demographic and facial-characteristic data without any stated consent, minimization, or handling guarantees.

External Transmission

Medium
Category
Data Exfiltration
Content
The queue endpoint returns a `run_id`. Use this to poll status:

```
GET https://api.comfydeploy.com/api/run/{run_id}
```

Poll until `status` is `"success"`, then extract the output image URL from the response.
Confidence
80% confidence
Finding
The polling endpoint indicates that run identifiers and output retrieval also depend on the external ComfyDeploy service, extending exposure beyond initial submission to result handling. This broadens the privacy surface because generated portraits and associated metadata may be stored or retrievable externally, yet the skill provides no warning or safeguards around retention, access, or result URLs.

Static analysis

No suspicious patterns detected.