# Task Delegation > Patterns for delegating tasks to specialized agents Use when this capability is needed. - Skill: `tomevault-io/task-delegation-3` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/task-delegation-3` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/task-delegation-3/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Productivity - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/task-delegation-3 --- # Task Delegation Skill ## When to Delegate Delegate when: - Task requires specialized expertise (testing, security, documentation) - Context is getting bloated (>70% usage) - Task can run in parallel with other work - Fresh perspective needed Don't delegate when: - Task is trivial (<5 minutes) - Heavy context sharing required - Sequential dependency on current work ## Delegation Patterns ### Sequential Delegation ```json { "mode": "sequential", "tasks": [ { "agent": "planner", "task": "Create plan" }, { "agent": "executor", "task": "Implement plan" }, { "agent": "verifier", "task": "Verify implementation" } ] } ``` ### Parallel Delegation ```json { "mode": "parallel", "max_concurrent": 3, "tasks": [ { "agent": "executor", "task": "Implement feature A" }, { "agent": "executor", "task": "Implement feature B" }, { "agent": "tester", "task": "Write tests" } ] } ``` ### Background Delegation ```json { "mode": "background", "agent": "researcher", "task": "Research best practices", "notify_on_complete": true } ``` ## Agent Selection | Task Type | Agent | Model Tier | |-----------|-------|------------| | Planning | goop-planner | quality | | Implementation | goop-executor-{tier} | balanced | | Verification | goop-verifier | quality | | Research | goop-researcher | balanced | | Documentation | goop-writer | budget | | Testing | goop-tester | balanced | | Debugging | goop-debugger | quality | | Security | goop-verifier | quality | ## Context Handoff When delegating, pass: 1. **Essential state:** Current phase, spec, todos 2. **Relevant files:** Only files the agent needs 3. **Recent decisions:** Last 3-5 ADL entries 4. **Constraints:** Boundaries, deadlines, blockers Don't pass: - Full conversation history - Verbose logs - Unrelated file contents - Completed task details ## Direct Task Delegation (CRITICAL) Delegation in GoopSpec uses the native **`task` tool** with rich, context-aware prompts constructed by the orchestrator. ### Prompt Construction Requirements Every delegation prompt MUST include: 1. **Task Intent** - What to build and why 2. **Project Context** - Stack, wave, existing patterns 3. **Constraints** - Boundaries and requirements 4. **Verification** - Commands to prove completion 5. **Expected Output** - Specific deliverables ### Example Delegation ```typescript task({ subagent_type: "goop-executor-high", description: "Implement user authentication", prompt: ` ## TASK Implement JWT-based user authentication with login/logout endpoints. ## PROJECT CONTEXT - Stack: Next.js 14 + NextAuth - Wave 2, Task 3 from BLUEPRINT.md - Follow existing patterns in src/auth/ - Use jose library for JWT (already in dependencies) ## CONSTRAINTS - Must support OAuth providers (Google, GitHub) - Token expiry: 24 hours with refresh rotation - Use existing session management in src/session/ ## VERIFICATION - Run: bun test src/auth/ - Manual: Test login/logout flow in browser - Check: No hardcoded secrets or credentials ## EXPECTED OUTPUT - src/auth/service.ts - JWT generation and validation - src/auth/middleware.ts - Route protection middleware - src/auth/types.ts - Auth type definitions - Atomic commit with verification evidence ` }) ``` ### When to Delegate | Situation | Use Delegation | |-----------|----------------| | Complex implementation | Yes - use appropriate executor tier | | Multi-file changes | Yes - include full context | | Architecture-sensitive | Yes - use goop-executor-high | | Simple config update | Optional - can be done directly | | Quick exploration | Optional - depends on context needs | ### Available subagent_types | subagent_type | Use For | |---------------|---------| | `goop-executor-low` | Simple implementation, config updates, mechanical fixes | | `goop-executor-medium` | Business logic, refactors, and standard implementation tasks | | `goop-executor-high` | Complex implementation, architecture-sensitive changes, critical code paths | | `goop-executor-frontend` | UI/UX implementation, styling, responsive frontend work | | `goop-explorer` | Fast codebase mapping, pattern detection | | `goop-researcher` | Deep domain research, technology evaluation | | `goop-planner` | Architecture design, blueprint creation | | `goop-verifier` | Verification against spec, security audit | | `goop-debugger` | Bug investigation, scientific debugging | | `goop-tester` | Test writing, coverage analysis | | `goop-designer` | UI/UX design, component architecture | | `goop-writer` | Documentation, technical writing | | `goop-librarian` | Code/docs search, information retrieval | | `general` | Fallback for any task | ### Common Mistakes | Mistake | Problem | Fix | |---------|---------|-----| | Vague prompt | Agent lacks context to succeed | Include all 5 required sections | | Wrong agent tier | Quality/speed mismatch | Match agent to task complexity | | Missing verification | Can't prove completion | Always specify verification commands | | No project context | Agent guesses patterns | Include stack, wave, existing patterns | | Using `delegate` tool | Different async system | Use `task` for GoopSpec agents | ### Full Example ```typescript task({ subagent_type: "goop-executor-high", description: "Implement password reset flow", prompt: ` ## TASK Implement password reset flow with email verification. ## PROJECT CONTEXT - Stack: Next.js 14, Prisma, Resend - Wave 2, Task 3 from BLUEPRINT.md - Follow existing auth patterns in src/auth/ - Email templates in src/templates/ ## CONSTRAINTS - Reset token expires in 1 hour - One-time use tokens only - Rate limit: 3 requests per hour per email - Must log all reset attempts ## VERIFICATION - Run: bun test src/auth/reset.test.ts - Manual: Test full reset flow with real email - Check: Token invalidation after use ## EXPECTED OUTPUT - src/auth/reset.ts - Reset logic - src/api/auth/reset.ts - API endpoint - src/templates/reset-email.tsx - Email template - Atomic commit with test evidence ` }) ``` ## Error Handling If delegated task fails: 1. Check error type (timeout, crash, assertion) 2. Save partial progress as checkpoint 3. Decide: retry, reassign, or escalate 4. Log failure to ADL if significant ## Best Practices 1. **Clear instructions:** Specific, unambiguous task descriptions 2. **Scoped context:** Only relevant information 3. **Defined success:** Clear verification criteria 4. **Timeout limits:** Set reasonable time bounds 5. **Progress tracking:** Monitor via todos/checkpoints --- > Converted and distributed by [TomeVault](https://tomevault.io/claim/hffmnnj) — claim your Tome and manage your conversions.