space-chart-image
Pass
Audited by Gen Agent Trust Hub on Aug 5, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads generated image binary data from
https://api.labnana.comthrough thegenerate_chart.pyscript when fallback mode is active. - [DATA_EXFILTRATION]: User-supplied chart descriptions, intent, and content are transmitted to an external service at
https://api.labnana.com/openapi/v1/images/generationto facilitate the generation of visual assets. - [COMMAND_EXECUTION]: The skill instructions in
SKILL.mdinvoke a local Python scriptscripts/generate_chart.pyvia shell command to handle image generation when internal environment models are unavailable. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted user data to construct prompts for the AI image generator.
- Ingestion points: User-provided descriptions and specific chart details extracted in the '收集内容' (Collect Content) phase as described in
SKILL.md. - Boundary markers: The skill lacks explicit delimiters or instructions to the image generator to ignore potential commands embedded within the user-provided chart data.
- Capability inventory: The skill possesses capabilities for network communication (API calls) and file system writes (saving PNG files to the user's directory) via
scripts/generate_chart.py. - Sanitization: There is no evidence of input validation or sanitization to prevent adversarial instructions in user input from manipulating the image generation output.
Audit Metadata