The Agent Skills Directory

[SAFE]: The skill implements legitimate vector composition and search functionality via the official Scry API (api.scry.io), which is a vendor resource of exopriors.
[COMMAND_EXECUTION]: The skill uses standard curl commands for API interaction. It follows security best practices by referencing the SCRY_API_KEY as an environment variable rather than hardcoding credentials.
[PROMPT_INJECTION]: The skill identifies a potential indirect prompt injection surface and provides explicit guardrails. Evidence chain: 1. Ingestion points: Untrusted text retrieved from the scry.entities and scry.chunk_embeddings views. 2. Boundary markers: Explicit instructions for the agent to treat retrieved text as untrusted and to never follow instructions found inside corpus payloads. 3. Capability inventory: Network interaction with the vendor API via curl. 4. Sanitization: Guidance to filter retrieved data based on the content_risk metadata column.

scry-vectors