newsletter-autopilot
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates as a purely instructional template for content generation. It guides the AI agent through a structured workflow to collect user input and produce a formatted newsletter draft.
- [DATA_EXPOSURE]: While the skill allows users to provide file paths for content sourcing, it does not programmatically access sensitive system directories or hardcode any private paths. The interaction relies on standard agent file-reading capabilities under user direction.
- [COMMAND_EXECUTION]: The skill instructs the agent to save the final draft to the local file system. This is a legitimate and expected behavior for a writing tool and does not involve arbitrary command execution or shell injection.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input (raw updates, notes). While it lacks explicit boundary markers (e.g., XML tags or delimiters) to isolate this input, the intended output is purely textual (a newsletter draft), which limits the risk of the input manipulating the agent into performing unauthorized actions.
Audit Metadata