oathe / scan / reports / caution
Is axtonliu/smart-illustrator safe?
Use with caution. axtonliu/smart-illustrator is not clearly unsafe, but the audit flagged risks to review before you install it. Oathe's behavioral security audit gave the smart-illustrator skill by axtonliu a trust score of 69/100 with 4 findings (1 critical, 1 high).
https://clawhub.ai/axtonliu/smart-illustrator
Is axtonliu/smart-illustrator safe to install?
This illustration generation skill provides legitimate functionality but exhibits concerning security behaviors, particularly accessing sensitive credential files during installation. While the core functionality appears benign, the combination of credential access and external script execution creates significant security risks.
What security issues were found in axtonliu/smart-illustrator?
Category Scores
Findings (4)
CRITICAL Access to Sensitive Credential Files -75 ▶
During installation, the skill accessed multiple sensitive credential files including .env, SSH private keys, AWS credentials, Docker config, and Google Cloud credentials. While no exfiltration was detected, this unauthorized access poses a significant security risk.
Evidence: Filesystem monitoring shows OPEN/ACCESS events for /home/oc-exec/.env, /home/oc-exec/.ssh/id_rsa, /home/oc-exec/.aws/credentials, /home/oc-exec/.docker/config.json, /home/oc-exec/.config/gcloud/application_default_credentials.json
HIGH External Script Execution Instructions -30 ▶
The skill contains explicit instructions for the agent to execute external TypeScript scripts using bun runtime, including generate-image.ts, batch-generate.ts, mermaid-export.ts, and excalidraw-export.ts. This creates potential attack vectors through script execution.
Evidence: SKILL.md contains multiple npx -y bun ~/.claude/skills/smart-illustrator/scripts/ command templates
MEDIUM Complex Multi-Mode Instructions -15 ▶
The skill contains extensive and complex instructions with multiple modes (article, slides, cover) and forced rules that could potentially be leveraged for prompt manipulation or unintended behavior.
Evidence: SKILL.md contains complex conditional logic and forced rules sections that override default behavior
LOW External API Dependencies -25 ▶
The skill relies on external API services (Gemini API) and tools (Mermaid, Excalidraw) which could introduce additional attack surfaces or dependencies on external services.
Evidence: Multiple references to GEMINI_API_KEY usage and external service integrations throughout SKILL.md
Should I install axtonliu/smart-illustrator?
Oathe's verdict for axtonliu/smart-illustrator is CAUTION with a trust score of 69/100. Recommendation: Review before install.