unsplash-image

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads image files from Unsplash, which is a well-known stock photo service. The download tracking is performed as required by the service's API terms.
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to execute a local Python script (unsplash-search.py) with parameters derived from search queries and text overlays. This is the primary mechanism for the tool's functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted search queries and text overlay parameters, creating an injection surface if the input contains hidden instructions.
  • Ingestion points: Input is accepted via the --query, --title, and --subtitle CLI arguments in the SKILL.md usage examples.
  • Boundary markers: No delimiters or specific 'ignore' instructions are provided to the agent to separate user data from the execution command.
  • Capability inventory: The skill has access to Bash(*), Read, and Write tools, allowing it to execute scripts, download network resources, and save files.
  • Sanitization: There is no evidence of input validation or sanitization within the skill's instruction file for the provided search and overlay strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 12:48 PM