ClawHub
Back to skill

Security audit

Process-Diagram-Pro

Security checks for vulnerabilities and agentic risk

Overview

This is primarily a local process-diagram generator, but its optional browser-automation documentation should be used cautiously.

Install the Python dependencies in a virtual environment if possible. Treat agent-browser as optional: use it only when you explicitly need web research, avoid logged-in or sensitive pages, and review any navigation, clicks, form fills, screenshots, or text extraction before allowing them.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (6)

Context-Inappropriate Capability

Medium
Confidence
96% confidence
Finding
The skill is presented as a diagram generator, but later adds browser automation capabilities that can open arbitrary URLs, click elements, fill forms, take screenshots, and extract page text. This materially expands the skill's effective permissions and can lead to unintended external browsing, data exfiltration from visited pages, or unsafe interaction with third-party sites, especially when framed as normal troubleshooting guidance.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger language says users can describe needs directly and the AI will automatically invoke the skill, with examples that are broad and generic. Overly broad auto-invocation increases the chance the skill runs in contexts the user did not intend, which is more dangerous here because the same file also documents unrelated browsing behavior.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The markdown gives concrete commands for visiting external sites, interacting with page elements, capturing screenshots, and retrieving page text, but provides no warnings about sensitive content, credential exposure, or local artifact creation. In practice, this normalizes risky browser automation and could cause confidential data from documentation portals or arbitrary websites to be copied, stored, or acted upon without informed consent.

Unpinned Dependencies

Low
Category
Supply Chain
Content
matplotlib>=3.7.0
pillow>=10.0.0
numpy>=1.24.0
Confidence
95% confidence
Finding
The dependency is specified with a lower bound only, which allows newer, unreviewed versions to be installed over time and makes builds non-reproducible. This increases supply-chain risk and can unexpectedly introduce breaking changes or vulnerable releases into the environment.

Unpinned Dependencies

Low
Category
Supply Chain
Content
matplotlib>=3.7.0
pillow>=10.0.0
numpy>=1.24.0
Confidence
97% confidence
Finding
Pillow is unpinned, so any future version satisfying the minimum constraint may be installed, including releases with security regressions or incompatible behavior. Because this package processes image data and has a history of memory-safety and parsing issues, leaving it unpinned raises practical supply-chain and exposure risk.

Unpinned Dependencies

Low
Category
Supply Chain
Content
matplotlib>=3.7.0
pillow>=10.0.0
numpy>=1.24.0
Confidence
95% confidence
Finding
NumPy is declared with a minimum version only, which permits drift to any later release and prevents reproducible installs. This can pull in versions with security defects or ABI/API changes that affect reliability and safety of the skill environment.

Static analysis

No suspicious patterns detected.