fusion-framework-feature-toggling

Pass

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to the ingestion of untrusted user data.
  • Ingestion points: User-provided goals, descriptions, and code context are collected for processing as described in SKILL.md and assets/offline-prompt-template.md.
  • Boundary markers: The instructions do not define explicit delimiters (such as XML tags or triple backticks) to isolate user-provided text from agent instructions when calling search tools.
  • Capability inventory: The skill utilizes mcp_fusion_search_framework and mcp_fusion_search_docs to query external documentation based on user input.
  • Sanitization: There is no evidence of validation or escaping for the user-provided code snippets before they are incorporated into the agent's context or tool queries.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 16, 2026, 09:24 AM