bmad-loop-resolve

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes content from externally provided files which could be manipulated to influence agent behavior through embedded instructions.
  • Ingestion points: The agent reads operational context from a JSON file path in $BMAD_LOOP_RESOLVE_CONTEXT, parses frozen specification Markdown files, and evaluates implementation artifacts like patch files and sentinel records.
  • Boundary markers: The skill instructs the agent to treat content within specific markers like as authoritative, though it does not implement formal validation for these boundaries.
  • Capability inventory: The agent is permitted to read and modify Markdown files and write a JSON resolution marker to the local file system. The skill explicitly forbids implementing feature code, running tests, or making commits, which limits the potential impact of an injection.
  • Sanitization: No specific routines for escaping or sanitizing the content of the processed files are defined in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 06:32 AM
Security Audit — agent-trust-hub — bmad-loop-resolve