image
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from local marketing context files, which could contain malicious instructions.
- Ingestion points: Instructs the agent to read
.agents/product-marketing.mdor similar files (SKILL.md). - Boundary markers: Absent.
- Capability inventory: Includes shell commands for image processing and
curl(SKILL.md). - Sanitization: Absent.
- [COMMAND_EXECUTION]: The skill provides example shell commands for image conversion and optimization.
- Evidence: Commands for
cwebp,mogrify,jpegoptim, and acurlpipe togrepare provided in SKILL.md for auditing and processing images. - [EXTERNAL_DOWNLOADS]: The skill references and links to several external AI image generation services and their APIs.
- Evidence: Links to official APIs for Gemini, Flux, Ideogram, OpenAI, and Recraft. These are well-known technology services and are used for the skill's primary purpose.
Audit Metadata