poster-design-generation
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes external data which could potentially contain malicious instructions. -- Ingestion points: The skill utilizes the 'image_urls' parameter in its API requests to process external images and supports multi-turn conversations via 'session_id' (SKILL.md). -- Boundary markers: There are no explicit delimiters or warnings provided to the agent to ignore instructions embedded within the ingested data. -- Capability inventory: The skill performs network operations via 'curl' commands to interact with the 'sense.eachlabs.run' API (SKILL.md). -- Sanitization: No mechanisms for sanitizing or validating the content of the images or session history are described.
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute shell commands to interact with an external API service. -- Evidence: Multiple examples in 'SKILL.md' guide the agent to use 'curl' for POST requests to 'https://sense.eachlabs.run/chat' using an API key stored in an environment variable.
Audit Metadata