writing-in-owner-voice
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or dangerous commands were detected. The skill consists of stylistic guidelines and a phrase bank intended to improve the naturalness of the agent's communication.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, untrusted data such as user-provided drafts, Slack messages, and project summaries. This creates a surface for indirect prompt injection where instructions hidden in the data could attempt to influence the agent's behavior. However, this is assessed as safe because the skill does not grant the agent any capabilities (such as network access, file system modifications, or tool execution) that could be abused through such an injection.
- Ingestion points: User-supplied drafts, standup updates, and PR content referenced in SKILL.md.
- Boundary markers: Not specified in the instructions.
- Capability inventory: None. The skill documentation explicitly states it 'Needs no tools' and contains no scripts or executable content.
- Sanitization: Not present.
Audit Metadata