design-to-code

Warn

Audited by Gen Agent Trust Hub on Feb 26, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on several shell commands for environment setup and script execution, including directory creation and dependency installation via pnpm.\n- [EXTERNAL_DOWNLOADS]: The configuration in scripts/package.json uses the 'latest' tag for the 'coderio' dependency, which results in the automatic download and execution of the newest version of an external package from a non-trusted source during the setup phase.\n- [CREDENTIALS_UNSAFE]: The documentation instructs users to provide their Figma Personal Access Token as a plaintext command-line argument to the fetch-figma command, potentially exposing it in shell history or process listings.\n- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through its handling of external Figma data.\n
  • Ingestion points: Document metadata and text content are retrieved from the Figma API by the scripts/coderio-skill.mjs script.\n
  • Boundary markers: The prompt generation logic does not utilize delimiters or specific instructions to isolate or ignore potentially malicious content within the fetched design data.\n
  • Capability inventory: The skill possesses the capability to write files to the local filesystem (e.g., src/ and scripts/ directories) using the save-code function.\n
  • Sanitization: There is no evidence of sanitization or validation of the data extracted from the Figma document before it is interpolated into LLM prompts.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 26, 2026, 12:56 AM