improve-human-writing-guide

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes several shell commands to compile LaTeX documents and process PDF files. This includes latexmk for compilation, pdftoppm for rendering pages to images, pdfinfo for metadata extraction, and file management commands like mkdir, cp, and cmp. These operations are central to the skill's purpose of building and verifying writing guides. It also invokes a repository-specific validation script scripts/validate-repo.py.- [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes external data, specifically a user-provided 'corpus' and 'target guides,' to match voice and style. This ingestion point presents an attack surface where malicious instructions in the source text could attempt to influence the agent's behavior.
  • Ingestion points: The skill reads a 'target guide' and 'representative source artifacts from the user's corpus' as described in SKILL.md.
  • Boundary markers: No explicit markers or warnings are used when processing these external artifacts.
  • Capability inventory: The skill has the ability to execute shell commands (latexmk, pdftoppm, cp, cmp, mkdir, pdfinfo) and run a repository validation script (scripts/validate-repo.py).
  • Sanitization: No sanitization or filtering of the ingested content is performed before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 12:50 AM