paper-writing-bench

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes external, potentially untrusted research paper content through LLM prompts. This creates a surface for indirect prompt injection where malicious instructions embedded in a paper could attempt to override the agent's behavior. The skill mitigates this by using [PAPER CONTENT] boundary markers.
  • Ingestion points: references/sparse-idea-prompt.md, references/dense-idea-prompt.md, and references/experimental-log-prompt.md via the {paper_content} placeholder.
  • Boundary markers: Present (uses [PAPER CONTENT] and [END PAPER CONTENT]).
  • Capability inventory: Generates markdown files via LLM calls; relies on the host environment's PDF extraction tools.
  • Sanitization: No explicit sanitization of the input text is performed before interpolation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 02:00 PM