doc-pipeline

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references official resources hosted on the author's GitHub repository ('claude-office-skills'). This is standard practice for distributing skill-related documentation and templates.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a pipeline architecture designed to process untrusted external data sources such as PDFs, spreadsheets, and images.
  • Ingestion points: The pipeline DSL (pipeline.yaml) and Python implementation accept external files as input data for extraction and AI analysis stages.
  • Boundary markers: The provided DSL examples do not explicitly define delimiters or specific 'ignore instructions' markers for the processed text.
  • Capability inventory: The skill leverages libraries for file manipulation (python-docx, openpyxl) and invokes AI models for analysis and summarization.
  • Sanitization: No explicit sanitization or filtering logic is present in the provided architectural examples, which is consistent with the framework-oriented nature of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 08:53 PM
Security Audit — agent-trust-hub — doc-pipeline