dispatch-orchestration

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill orchestrates a pipeline that processes untrusted data from external platforms to generate and execute tasks.
  • Ingestion points: The skill fetches work items from GitHub Issues and Azure DevOps as mentioned in SKILL.md.
  • Boundary markers: There are no explicit instructions or delimiters provided to ensure the agent ignores malicious instructions embedded within the fetched issue descriptions.
  • Capability inventory: The Dispatch tool possesses high-privilege capabilities including code generation, local file system operations, and the ability to commit code and open pull requests.
  • Sanitization: The documentation does not describe any sanitization or validation steps for external issue content before it is used to generate task specifications.
  • [COMMAND_EXECUTION]: The skill provides comprehensive instructions for executing the dispatch CLI and its various subcommands to manage task execution, planning, and configuration. These operations are essential to the skill's purpose but involve significant interaction with the host environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 02:23 PM