bmad-loop-resolve
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources which could contain malicious instructions targeting the agent's behavior.
- Ingestion points: The agent is instructed to read files from the environment variable
$BMAD_LOOP_RESOLVE_CONTEXT, thespec_filepath,stories.yaml,SPEC.md, and patch files referenced in escalation details. - Boundary markers: There are no instructions to use specific delimiters or protective headers when reading or interpolating the content of these external files into the agent's context.
- Capability inventory: The skill has file read/write permissions within the project directory (specifically to update implementation specs and write a
resolution.jsonmarker) and is authorized to invoke downstream skills likebmad-spec,bmad-correct-course, andbmad-agent-po. - Sanitization: The instructions lack requirements for sanitizing, filtering, or escaping the content of the ingested files before using them to inform the resolution or update other project artifacts.
Audit Metadata