write-background

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: LOWNO_CODEPROMPT_INJECTION
Full Analysis
  • [NO_CODE] (SAFE): The skill is comprised entirely of natural language instructions and templates, with no scripts or command-line operations found.
  • [PROMPT_INJECTION] (LOW): The skill ingests untrusted content from research files. Ingestion points: .research/literature/.md and .research/literature/.bib. Boundary markers: None. Capability inventory: Limited to drafting text in manuscript/background.md. Sanitization: None. Impact is limited to content manipulation within the generated manuscript draft.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 17, 2026, 08:07 AM