task-coordinator

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill defines structured patterns for managing complex workflows using a built-in task system. This involves standard operations like creating tasks, updating status, and tracking dependencies.
  • [PROMPT_INJECTION]: The orchestration logic defines a surface for indirect prompt injection through its use of a local ledger file and dynamic task prompts, though it is used for legitimate workflow management.
  • Ingestion points: Workflow state is read from .agents/workflow-ledger.yml.
  • Boundary markers: No explicit delimiters or boundary markers are defined in the orchestration templates.
  • Capability inventory: The skill uses native Task functions (TaskCreate, TaskUpdate, Task()) for subagent invocation, state management, and file writing to the ledger path.
  • Sanitization: No explicit sanitization of ledger content is shown, as the data is primarily used for internal workflow metadata and status tracking.
  • [SAFE]: The skill utilizes platform-specific configuration commands (e.g., /majestic:config) and environment variables for managing task tracking settings and cross-session persistence, which are consistent with the intended functionality of the orchestration tool.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 02:36 PM