technical-writing-reviser
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted technical drafts which are then used to build internal revision briefs for sub-agent workers, creating an indirect prompt injection surface.
- Ingestion points: Technical artifacts and document drafts provided by the user as described in SKILL.md.
- Boundary markers: Absent; the instructions for constructing worker prompts do not specify the use of isolation markers or delimiters for the untrusted content.
- Capability inventory: The skill is limited to text revision and does not request or use tools for shell execution, network access, or file system modification.
- Sanitization: No specific sanitization or filtering of input text is defined; however, the skill's narrow focus on evidence-based technical claims and its disjoint worker ownership model provide logical constraints.
- [SAFE]: No malicious code, obfuscated content, or unauthorized command executions were found. The skill adheres to technical writing best practices and provides clear guidelines for maintaining technical accuracy.
Audit Metadata