dmux-workflows

Pass

Audited by Gen Agent Trust Hub on Aug 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references an external GitHub repository (github.com/standardagents/dmux) for tool installation. While it advises a manual review, the source is not an inherently trusted platform vendor.- [COMMAND_EXECUTION]: The orchestration helper script (scripts/orchestrate-worktrees.js) executes arbitrary shell commands specified in the launcherCommand field of a JSON configuration.- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through its orchestration plan processing:
  • Ingestion points: Reads task descriptions and launcher commands from plan.json.
  • Boundary markers: None identified in the provided documentation.
  • Capability inventory: Executes shell commands, creates git worktrees, and writes orchestration metadata to the local file system.
  • Sanitization: No evidence of input validation or command escaping for the JSON-provided fields.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 5, 2026, 03:18 PM