The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill includes a hardcoded Algolia API key (3e42dbd4f895fe93ff5cf40d860c4a85) and Application ID (PLSDUOL1CA) in SKILL.md for executing documentation searches.
[INDIRECT_PROMPT_INJECTION]: The skill fetches documentation content from external sources, creating a surface for processing untrusted instructions.
Ingestion points: Search metadata from algolia.net and Markdown documentation fetched from remotion.dev.
Boundary markers: No delimiters are specified to isolate fetched content from agent instructions.
Capability inventory: The skill instructs the agent to make network requests (HTTP POST and GET) and act upon retrieved data.
Sanitization: No sanitization or validation of external content is described.

remotion-docs