The Agent Skills Directory

[SAFE]: No malicious patterns, prompt injections, or obfuscation techniques were identified in the skill instructions or associated scripts.
[COMMAND_EXECUTION]: The skill uses Deno scripts for design automation. These scripts are designed to run with --allow-read and --allow-write permissions, which are strictly necessary for the stated purpose of auditing project files and writing generated design artifacts to the file system.
[INDIRECT_PROMPT_INJECTION]: The scripts ingest external data such as CSS, TSX, and JSON specification files.
Ingestion points: External content is read in analyze-styles.ts, extract-tokens.ts, analyze-accessibility.ts, and generate-tokens.ts.
Boundary markers: None explicitly defined in the markdown instructions.
Capability inventory: Limited to file system read/write operations; no network access or subprocess spawning capabilities are included in the scripts.
Sanitization: Input content is parsed using regular expressions or JSON.parse without any dynamic execution of the provided data, effectively mitigating the risk of command injection or malicious code execution via data ingestion.
[EXTERNAL_DOWNLOADS]: The scripts import modules from Deno's JSR registry using versioned imports (e.g., jsr:@std/cli@1.0.9). These are well-known and reputable sources within the Deno ecosystem.

frontend-design