The Agent Skills Directory

[SAFE]: The skill serves as a guide for developers to implement efficient API patterns such as caching, batching, and server-side filtering to minimize Canva Connect API overhead.
[SAFE]: Code snippets are illustrative TypeScript examples for architectural patterns and do not contain executable commands, remote downloads, or credential harvesting logic.
[INDIRECT_PROMPT_INJECTION]: The skill describes processes that ingest external data from the Canva Connect API (e.g., design titles, metadata).
Ingestion points: findDesignByTitle and exportIfNeeded snippets in SKILL.md process data returned from the Canva API.
Boundary markers: Not explicitly present in the provided logic snippets.
Capability inventory: The skill is restricted to Read and Grep tools as per the YAML frontmatter, minimizing the risk of harmful actions if malicious data is encountered.
Sanitization: Standard caching and tracking logic is shown; explicit sanitization of API-returned strings is not included in the snippets.
[SAFE]: External links point to official Canva documentation and pricing pages (canva.com, canva.dev), which are well-known and trusted services.

canva-cost-tuning