seedream
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch a remote markdown file (
https://www.pixazo.ai/models/seedream.md) for "exact parameter names, accepted values, or aren't sure about a field." This content is intended to guide the agent's logic at runtime. - Ingestion points:
SKILL.mdinstructions guide the agent to load content from an external URL. - Boundary markers: The skill does not provide delimiters or instructions to ignore embedded commands within the fetched document.
- Capability inventory: The skill utilizes network operations (
curl,requests,fetch) and file system access (writing to~/.pixazo/api-key). - Sanitization: No sanitization or validation of the remote markdown content is described.
- [COMMAND_EXECUTION]: The skill includes a command in the "Related Pixazo skills" section (
npx skills add Pixazo-AI/skills --skill '*') which suggests the execution of external packages. - [EXTERNAL_DOWNLOADS]: The skill fetches operational guidance and technical specifications from the vendor's website (
pixazo.ai) during the execution flow.
Audit Metadata