pp-visit-detroit-blog
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires installing a CLI tool from the developer's NPM package (@mvanhorn/printing-press-library) or GitHub repository (github.com/mvanhorn). These are vendor-owned resources provided to enable the skill's capabilities.
- [COMMAND_EXECUTION]: The skill operates by executing the visit-detroit-blog-pp-cli binary via Bash to handle queries, content retrieval, and reporting.
- [DATA_EXFILTRATION]: The CLI includes a --deliver webhook: flag that allows the user to POST command results to external servers. It also provides a feedback mechanism that can be configured to send data to a remote endpoint via environment variables.
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it ingests and processes editorial content from a third-party website.
- Ingestion points: Articles are fetched from the Visit Detroit blog during sync and retrieval operations.
- Boundary markers: The instructions do not define delimiters or specific guidance to the agent to treat fetched blog text as untrusted data.
- Capability inventory: The tool includes file system write access (--output, file:) and network request capabilities (webhook delivery).
- Sanitization: There is no description of sanitization or filtering applied to the external blog content before processing.
Audit Metadata