skills/xoai/sage/pack-source-process/Gen Agent Trust Hub

pack-source-process

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted external data (GitHub issues, blogs, migration guides) and user-provided project context, creating a surface for indirect prompt injection attacks.
  • Ingestion points: External research sources and project context documents are read during 'Step 1: Source Prioritization' and 'Step 4: Project Overlay Processing'.
  • Boundary markers: The skill does not specify the use of delimiters or 'ignore embedded instructions' warnings for the data being analyzed.
  • Capability inventory: The skill uses the agent's ability to read external content and write processed results to the '.sage/pack-build/sources.md' file.
  • Sanitization: No sanitization, escaping, or strict validation is instructed for the content extracted from the external sources.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 01:31 AM
Security Audit — agent-trust-hub — pack-source-process