space-chart-image
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data and uses it to construct prompts for an external image generation model.
- Ingestion points: User input describing chart details is collected as defined in the SKILL.md workflow.
- Boundary markers: There are no delimiters or instructions to ignore embedded commands when processing user input in references/chart-prompts.md or scripts/generate_chart.py.
- Capability inventory: The skill writes generated images to the local file system using scripts/generate_chart.py.
- Sanitization: User input is not sanitized or escaped before being included in the prompt for the image generation API.
- [EXTERNAL_DOWNLOADS]: The skill performs network requests to an external endpoint (api.labnana.com) to generate chart images, which is a non-whitelisted domain.
Audit Metadata