pm-writing
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided text for rewriting or generating documents, which creates a surface for indirect prompt injection.
- Ingestion points: The skill accepts arbitrary text via the argument-hint and within its "Rewrite" and "Generate" modes as defined in
SKILL.md. - Boundary markers: The instructions do not define delimiters or specific system-level warnings to distinguish user-provided data from agent instructions.
- Capability inventory: The skill metadata does not restrict tool usage via
allowed-tools, meaning the agent may have access to its full suite of capabilities (such as file system access or shell execution) while processing potentially malicious user-provided text. - Sanitization: No sanitization, escaping, or validation of user input is specified in the instructions.
Audit Metadata