pp-nasa-images
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to install a CLI tool using
npx -y @mvanhorn/printing-press-libraryorgo install github.com/mvanhorn/printing-press-library/.... These resources are hosted on infrastructure associated with the skill's vendor. - [COMMAND_EXECUTION]: The skill operates by executing the
nasa-images-pp-clibinary through the agent's bash capabilities to perform media searches and downloads. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from NASA's public API, which could theoretically contain instructions designed to influence the agent. \n- Ingestion points: Data is ingested via the
captions fetch,metadata fetch, andmedia searchcommands. \n- Boundary markers: The skill instructions do not specify any delimiters or special handling to differentiate between data content and agent instructions. \n- Capability inventory: The skill environment allows for shell command execution, local file system writes, and network access. \n- Sanitization: There is no evidence of sanitization or filtering of the content retrieved from the NASA Image and Video Library. - [DATA_EXFILTRATION]: The
nasa-images-pp-clitool includes a--deliver webhook:<url>option. This feature allows the output of commands to be sent to a user-specified remote URL, which could be misused to exfiltrate data if a malicious destination is provided.
Audit Metadata