The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill uses an 'escape hatch' mechanism (Step 3-delta) that involves generating and executing JavaScript code using the figma_execute or use_figma tools to fetch missing data. While the instructions prescribe limits (e.g., read-only, under 50 lines), the agent has the capability to construct and execute arbitrary code based on its internal logic.
[INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it processes unvalidated external data and user-provided instructions to drive its reasoning and code generation.
Ingestion points: Reads data from {componentSlug}-_base.json and accepts optionalContext as a free-form user string.
Boundary markers: Lacks delimiters or specific instructions to isolate or ignore potentially malicious instructions embedded in the input data or context.
Capability inventory: Possesses tools for file system access (Read, Write) and external code execution (figma_execute, use_figma).
Sanitization: No validation or filtering is performed on the user context or JSON content before it is processed or used in script generation.

extract-api