writing-pre-publish-checklist
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-supplied writing, which could potentially contain malicious instructions intended to divert the agent from its checklist task. However, the impact is minimal as the skill lacks exploitable capabilities such as network access, file modification, or command execution. \n
- Ingestion points: The skill accepts a "Finished or near-finished piece of writing" as primary input. \n
- Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the input as untrusted data. \n
- Capability inventory: Analysis of SKILL.md and associated resources confirms no subprocess calls, network operations, or file-writing tools are utilized. \n
- Sanitization: No input sanitization or validation logic is present for the analyzed text. \n- [SAFE]: The skill uses a local JSON resource for evaluation criteria (resources/evaluators/rubric_pre_publish.json), which contains standard, non-executable metadata. No external downloads or remote code patterns were detected.
Audit Metadata