parallel-plan-feature

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted content from the codebase and user arguments.
  • Ingestion points: User-provided feature description ($ARGUMENTS) and project files (openspec/project.md, architecture.summary.json).
  • Boundary markers: The instructions do not define explicit delimiters or instructions to ignore embedded instructions in the ingested content.
  • Capability inventory: The skill can execute git commands, local scripts, and register resource claims via the coordinator.
  • Sanitization: No input sanitization or validation of the content being processed into proposals is specified.
  • [COMMAND_EXECUTION]: The skill executes local git commands (git pull, git status) and runs specific validation scripts (scripts/validate_work_packages.py, scripts/parallel_zones.py) within the project's repository. These operations are within the expected behavior of the skill.
  • [SAFE]: No remote code execution from unknown sources, hardcoded credentials, data exfiltration to external domains, or persistence mechanisms were detected. The scripts being executed are local vendor resources.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 10:38 PM