The Agent Skills Directory

[DATA_EXFILTRATION]: The skill performs network requests via curl to the official Figma API (api.figma.com) to verify authentication and connectivity. This targets a well-known service and adheres to security best practices by using environment variables (e.g., FIGMA_TOKEN) instead of hardcoded secrets. The documentation includes explicit warnings against logging or sharing personal access tokens.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external Figma metadata which presents an attack surface for indirect injection.
Ingestion points: Figma file keys and node identifiers provided by the user or extracted from Figma URLs as described in SKILL.md.
Boundary markers: The instructions recommend starting from a "sanitized error record" and utilizing a "disposable design fixture" for reproduction.
Capability inventory: The skill uses Bash(curl:*) to perform authenticated queries against external APIs.
Sanitization: The skill emphasizes the use of "sanitized results" and "scoped credentials" to mitigate the risk of accidental data exposure during diagnostics.

anima-common-errors