create-note
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows a legitimate orchestration pattern with no evidence of malicious intent, command execution, or network exfiltration.
- [DATA_EXFILTRATION]: The skill interacts with local files at
./newsletter/issues/[issue_name]/plan.mdand./newsletter/issues/[issue_name]/notes.md. This access is confined to the specific directory related to the user's content and does not involve sensitive system data or external network transmission. - [PROMPT_INJECTION]: The skill identifies an indirect prompt injection surface. 1. Ingestion points:
plan.mdis read to provide context for downstream tasks. 2. Boundary markers: Absent. 3. Capability inventory: The skill triggerscontent-strategy:researchandwriting:copywritinginvocations and performs file writes tonotes.md. 4. Sanitization: Absent. Given that the skill is intended to process user-provided planning documents within a local directory, this behavior is a functional requirement and does not present a high-risk vulnerability.
Audit Metadata