The Agent Skills Directory

[SAFE]: The skill is a standard API bridge that facilitates interaction with the find-scene.com service. It provides clear instructions for handling authentication tokens and defines specific, functional endpoints for movie-related tasks.
[PROMPT_INJECTION]: The skill identifies surfaces for indirect prompt injection by processing external subtitle data (SRT) and movie metadata. However, the risk is assessed as SAFE because the data originates from a curated video indexing service and the agent's capabilities are limited to performing network requests to the official API.
Ingestion points: Subtitle strings (srt field) and movie metadata returned by endpoints such as /api/search_phrase, /api/get_srt_entries_by_time_range, and /api/transcribe_by_time in SKILL.md.
Boundary markers: Not implemented.
Capability inventory: Network requests to api.find-scene.com as defined in the workflows and endpoint reference in SKILL.md.
Sanitization: No explicit sanitization or filtering is described in the skill instructions.

find-scene