Agent Delegation¶
Framework for delegating tasks to agents with a durable audit trail and human review gates.
When to delegate¶
Delegate a task to an agent when: - The work is well-scoped with clear acceptance criteria - The output is a concrete artifact (code, document, research summary) - Human review of the output is feasible before it's acted on
How to delegate¶
- Copy a template from
templates/that matches the task type - Fill in the scope, acceptance criteria, and output location
- Hand the filled template to the agent (paste into Claude Code or Cursor chat)
- When the agent completes, review the output against the acceptance criteria
- Append an entry to
audit-log.mdwith the outcome
Templates¶
| Template | Use for |
|---|---|
| task-delegation.md | General code, research, or documentation tasks |
Audit log¶
All completed delegations are recorded in audit-log.md. This is an append-only log — never edit past entries.
Review gates¶
- Sync review (default): You review output before it's saved or acted on
- Async with spot-check: Agent writes output to a draft location; you review at next available time
- Mark the review status in the audit log entry:
approved,revised, orrejected