scientific-writing

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses local subprocesses to automate figure generation and document compilation. These include calls to scripts/generate_schematic.py and scripts/generate_image.py for scientific visuals, and the xelatex engine for generating high-quality PDF reports.
  • [EXTERNAL_DOWNLOADS]: The skill integrates with research-lookup to retrieve scientific literature and data from external repositories.
  • [PROMPT_INJECTION]: There is a potential surface for indirect prompt injection due to the processing of untrusted data from external research papers.
  • Ingestion points: Data and literature retrieved from the research-lookup skill (SKILL.md).
  • Boundary markers: The skill implements a two-stage development process (outline creation followed by prose conversion), which provides some logical separation, although explicit delimiters for external content are not mentioned.
  • Capability inventory: Subprocess execution capabilities via Python and LaTeX.
  • Sanitization: No explicit validation or filtering logic for the retrieved research data is described in the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 05:50 PM