bmad-epic-pipeline-worktree

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is designed to programmatically invoke other skills and execute arbitrary shell commands defined in the workflow-steps.md configuration file. It manages Git worktrees and performs branch merges as part of its core logic.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing external data files to determine its execution flow.
  • Ingestion points: Reads task metadata and status information from _bmad-output/implementation-artifacts/sprint-status.yaml or docs/sprint/sprint-status.yaml.
  • Boundary markers: None identified. The skill does not use specific delimiters or safety instructions to prevent the agent from obeying instructions potentially embedded within the YAML data (e.g., in story names).
  • Capability inventory: Executes the /bmad-story-pipeline-worktree command, which involves filesystem modifications (worktrees), Git operations (merge, add, commit), and any shell commands specified in the workflow-steps.md pipeline config.
  • Sanitization: No validation or sanitization of the input data from the YAML files is performed before it is interpolated into subagent tasks or logical branching decisions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 07:41 AM