diagram-gen
3 findings (1 critical, 1 high, 1 low), mainly policy violation, data exfiltration, social engineering. Most recent scan 2026-02-06.
This skill presents CRITICAL and HIGH security risks primarily due to undisclosed external data transmission. While the skill's stated purpose (generating Mermaid diagrams) is legitimate, it conceals that user source code will be sent to OpenAI's API, requiring an API key that's only mentioned in fine print. The description uses social engineering tactics ('Zero config', 'Just works') to obscure the actual requirements and data handling behavior. The skill lacks proper metadata declarations (allowed-tools, compatibility) and transparency about its external dependencies. Most concerning is the potential for sensitive source code and credentials to be transmitted to third-party services without explicit user consent or prominent disclosure. The skill requires significant transparency improvements and proper security controls before it can be considered safe for use with potentially sensitive codebases.
Scanner's overall assessment (LLM analyzer)
Second scanner: NVIDIA SkillSpector
Independent static run, no language model, over the same files as the Cisco scan above. How the two scanners compare across the whole corpus: Two scanners, one corpus.
What this skill does
Generate Mermaid diagrams from your codebase. Use when you need architecture visualizations.
- Package
- 7 files, 53 KB
- SKILL.md
- 291 words, 12+ sections
- Code
- TypeScript (2)
SKILL.md covers: Diagram Generator · Quick Start · What It Does · Usage Examples · Generate flowchart · Class diagram · Sequence diagram of function calls · Entity relationship diagram · Best Practices · When to Use This · Part of the LXGIC Dev Toolkit · Requirements
Findings
CRITICAL Hardcoded External Dependency Requires API Key Without Disclosure Data Exfiltrationllm
The skill requires OPENAI_API_KEY environment variable but this is only mentioned in a requirements section, not prominently disclosed in the description or manifest. The skill will access and potentially transmit user code to OpenAI's API without explicit user consent in the main description. This constitutes unauthorized data exfiltration of potentially sensitive source code to a third-party service.
SKILL.md
Requirements section states: 'Needs OPENAI_API_KEY environment variable' and 'Reads your source files to understand the structure'. The skill sends user code to OpenAI API but the description only says 'Generate Mermaid diagrams from your codebase' without disclosing external API usage.HIGH Deceptive Description Conceals External API Dependency Social Engineeringllm
The skill description claims 'One command. Zero config. Just works' and 'No install needed. Just run with npx' but conceals that it requires OPENAI_API_KEY configuration and sends user code to external servers. This is social engineering through misleading marketing that obscures the actual data handling behavior.
SKILL.md
Description: 'One command. Zero config. Just works.' vs Reality: Requires OPENAI_API_KEY environment variable and transmits code to OpenAI. The 'zero config' claim is false and the external data transmission is not disclosed upfront.LOW Skill does not specify a license Policy Violationstatic
Skill manifest does not include a 'license' field. Specifying a license helps users understand usage terms.
SKILL.md
What is in the package
| File | Kind | Size |
|---|---|---|
README.md | doc | 718 B |
SKILL.md | doc | 2.1 KB |
package-lock.json | config | 46 KB |
package.json | config | 737 B |
tsconfig.json | config | 217 B |
src/cli.ts | script | 1.2 KB |
src/index.ts | script | 1.5 KB |
External hosts referenced
registry.npmjs.org github.com feross.org www.patreon.com lxgicstudios.com lxgicstudios.substack.com paypal.me x.com
Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.
Package contents recorded 2026-10-22. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.
What we checked
- Static analyzer. Pattern rules over SKILL.md and bundled files: known-bad commands, exfiltration URLs, secrets, obfuscation.
- Behavioral analyzer. Checks what the skill asks the agent to do against what its description claims it does.
- Llm analyzer. A language model reads the skill and reports intent-level issues the rules cannot express.
- Trigger analyzer. Looks for instructions that fire on conditions the user did not ask for.
- Meta analyzer. A second model pass that validates, deduplicates and re-scores the other analyzers' findings.
What we did not check
- Runtime behaviour. The skill was not executed; findings come from reading its files.
- Anything fetched at install or run time, including remote scripts, package dependencies, and URLs the skill tells the agent to open.
- Content the analyzers cannot read: images, binaries, and some non-English text.
- Known scanner bypass classes. See the methodology and limitations page.
Sources
- Repository: https://clawhub.ai/skills/diagram-gen
- Raw scan result: diagram-generator-scan.json
- Badge for your README:
- Request a re-scan