The Agent Skills Directory

[DATA_EXPOSURE]: The skill accesses configuration files such as .env, .env.local, and .figma.env to retrieve the FIGMA_ACCESS_TOKEN. This is a standard practice for local development tools. The script includes a maskToken function to ensure sensitive credentials are not printed in full to the console during configuration checks.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from the Figma REST API and user-provided URLs, creating a surface for indirect prompt injection.
Ingestion points: Data enters through the --url argument, the --urls-file argument (which reads local files), and the responses from api.figma.com.
Boundary markers: Output JSON files do not contain explicit delimiters or instructions for downstream agents to ignore embedded content.
Capability inventory: The script performs network GET requests and writes results to the local filesystem using fs.writeFileSync.
Sanitization: The tool uses a safeNodeName function to sanitize Figma node IDs (replacing non-alphanumeric characters with dashes), effectively preventing path traversal attacks when generating output directory structures.
[COMMAND_EXECUTION]: The skill provides CLI-based Node.js scripts. While it performs filesystem and network operations, it does not use shell execution functions like eval(), exec(), or spawn() with user-controlled input.

figma-node-fetcher