manage-themes

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill follows secure design principles for theme management, including centralized storage in a resolved workspace directory and structured auditing for accessibility. No malicious patterns or obfuscation techniques were detected.
  • [COMMAND_EXECUTION]: Invokes local python utilities such as generate-tokens-css.py, validate-theme-manifest.py, and import-claude-design-bundle.py. These scripts are part of the vendor's (cogni-work) official workspace scripts and are used for processing design tokens and verifying theme integrity.
  • [EXTERNAL_DOWNLOADS]: Connects to api.anthropic.com to import design bundles and uses browser automation to read external websites. These operations are intended for fetching design assets and extracting brand styles as part of the skill's primary function.
  • [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection because it ingests data from external websites and PowerPoint files. However, this is inherent to the theme extraction functionality, and the risk is limited by the structured visual analysis process prescribed in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 01:51 PM
Security Audit — agent-trust-hub — manage-themes