webflow-mcp:review-comments
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted user-generated content (Webflow comment threads) in Step 2 and Step 4. While the skill provides specific triage logic, there are no explicit boundary markers or instructions for the agent to ignore potentially malicious instructions embedded within the comment text itself (Indirect Prompt Injection surface).
- Ingestion points: Comment content fetched via
data_comments_tool > list_comment_threadsinSKILL.md. - Boundary markers: Absent. The skill does not instruct the agent to treat comment content as data only.
- Capability inventory: The skill has the ability to post replies back to the Webflow API (
data_comments_tool > create_reply) and perform local file system operations (mkdir,write_file). - Sanitization: None described for the comment content before it is processed or used in report generation.
- [COMMAND_EXECUTION]: The skill performs local file system operations including creating a directory (
mkdir comment-reviews) and writing markdown files. These are expected behaviors for a reporting tool but represent a standard command execution surface.
Audit Metadata