screen-spec-generator

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes shell commands (mkdir -p) to create and organize the directory structure for documentation and screenshots within the project's docs/screen_specs/ folder.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface (Category 8) due to its analysis of untrusted project files at runtime.
  • Ingestion points: The skill reads and parses CLAUDE.md, pubspec.yaml, and Flutter .dart source files to extract information about the application's UI and logic.
  • Boundary markers: No specific delimiters or "ignore" instructions are provided to the agent to prevent it from potentially following instructions embedded within code comments or string literals found in the analyzed files.
  • Capability inventory: The agent has the capability to read project files, create directories, and write new Markdown and HTML files.
  • Sanitization: There is no evidence of sanitization or filtering of the content extracted from the source code before it is interpolated into the final documentation templates.
  • [DYNAMIC_EXECUTION]: The skill generates a custom command file at .claude/commands/screen-spec.md. This serves as a persistent extension of the agent's capabilities, allowing it to re-execute the documentation generation logic using the localized templates and instructions provided by the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 10:42 PM