tmdb
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes movie metadata such as titles and overviews from the external TMDb API, which represents a potential attack surface for indirect prompt injection.
- Ingestion points: Untrusted data is retrieved from
api.themoviedb.orgwithin thesearch,movie,tv, andsuggestcommands inscripts/tmdb.py. - Boundary markers: The skill does not use specific delimiters to wrap external content when returning it to the agent context.
- Capability inventory: The skill's capabilities are limited to network requests to the TMDb API and writing to local files (
watchlist.json,preferences.json). It does not perform arbitrary command execution or access sensitive system files. - Sanitization: The script performs basic sanitization by truncating movie overviews to 200 characters.
Audit Metadata