The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill describes methods for plugins to ingest data from external sources and sync it to the Framer CMS or canvas. This ingestion of untrusted data, combined with capabilities like creating code files and modifying nodes, represents a standard indirect prompt injection attack surface. However, the skill provides sanitization examples (slug generation) and emphasizes permission-based access control to mitigate risks.
[DYNAMIC_EXECUTION]: The skill documents core SDK features for programmatic creation and modification of code files (framer.createCodeFile) and component instances. These are legitimate capabilities of the Framer platform, and the skill includes appropriate instructions for permission gating and handling code-file visibility safely.
[EXTERNAL_DOWNLOADS]: The documentation references standard developer resources and well-known services, such as official Framer documentation domains, esm.sh for module loading, and placeholders for OAuth workers. These references are consistent with the skill's purpose as a developer guide and follow industry best practices.

framer-plugins