Skills
Navigate
skills/mphinance/alpha-skills/edge-strategy-designer/Gen Agent Trust Hub

edge-strategy-designer

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from an external YAML file and incorporates it into generated strategy drafts.\n
  • Ingestion points: The load_concepts function in scripts/design_strategy_drafts.py reads data from a user-provided edge_concepts.yaml file.\n
  • Boundary markers: The output YAML files do not implement boundary markers or instructions for downstream agents to disregard embedded content.\n
  • Capability inventory: The script performs file system write operations via Path.write_text in scripts/design_strategy_drafts.py to save the generated drafts.\n
  • Sanitization: While the script uses sanitize_identifier for IDs, free-text fields such as thesis and representative_conditions are included in the output without specific filtering for malicious instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 07:50 AM