feedsmith
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references the feedsmith package on NPM and its documentation site v3.feedsmith.dev. It provides commands to install the library using bun add. These references are standard for the library's use and point to established resources.- [COMMAND_EXECUTION]: The instructions include command-line examples for package management (bun pm ls, npm view, bun info). These tools are used for verifying the environment and installing dependencies, which aligns with the skill's purpose.- [PROMPT_INJECTION]: The skill is designed to parse untrusted external data in RSS, Atom, and JSON formats, creating a surface for indirect prompt injection.
- Ingestion points: Data is ingested through the parseFeed function and specific parser APIs described in references/parsing.md.
- Boundary markers: There are no explicit instructions to the agent to use boundary markers or to ignore instructions embedded within the feed data it parses.
- Capability inventory: The skill has the capability to run package management commands.
- Sanitization: The documentation focuses on the structural parsing of the feeds and does not detail sanitization measures for the content itself.
Audit Metadata