Dispatching parallel agents

Dispatching Parallel Agents

Every dynamic worker prompt must include the resolved path to skills/using-superpowers/references/technical-writing-policy.md. Require the worker to read the policy before its first report.

Overview

You delegate tasks to specialized agents with isolated context. By precisely crafting their instructions and context, you ensure they stay focused and succeed at their task. They should never inherit your session's context or history — you construct exactly what they need. This also preserves your own context for coordination work.

When you have multiple independent problem domains—such as unrelated bug investigations or subsystem failures—executing them sequentially wastes time. Each domain can proceed in parallel when agents have isolated workspaces. Graph-plan tasks route to SDD instead.

Core principle: Dispatch one agent per independent problem domain. Let them work concurrently.

When to Use

digraph when_to_use {
    "Multiple failures?" [shape=diamond];
    "Are they independent?" [shape=diamond];
    "Single agent investigates all" [shape=box];
    "One agent per problem domain" [shape=box];
    "Can they work in parallel?" [shape=diamond];
    "Sequential agents" [shape=box];
    "Parallel dispatch" [shape=box];

    "Multiple failures?" -> "Are they independent?" [label="yes"];
    "Are they independent?" -> "Single agent investigates all" [label="no - related"];
    "Are they independent?" -> "Can they work in parallel?" [label="yes"];
    "Can they work in parallel?" -> "Parallel dispatch" [label="yes"];
    "Can they work in parallel?" -> "Sequential agents" [label="no - shared state"];
}

Use when:

• 3+ test files failing with different root causes
• Multiple subsystems broken independently
• Each problem can be understood without context from others
• No shared state between investigations
• Multiple independent subsystem changes (different files, different concerns)

Don't use when:

• Failures are related (fix one might fix others)
• Need to understand full system state
• Agents would interfere with each other

The Pattern

1. Identify Independent Domains

Group failures by what's broken:

• File A tests: Tool approval flow
• File B tests: Batch completion behavior
• File C tests: Abort functionality

Each domain is independent - fixing tool approval doesn't affect abort tests.

Before you fan out (orchestrator-only)

Worktrees isolate files, not assumptions — parallel agents on different files can still diverge on an un-prescribed shared decision (MAST FC2). Before dispatching:

1. Front-load shared decisions — list every decision ≥2 agents depend on (schemas, naming, interfaces, conventions); decide each once and write it verbatim into every agent prompt.
2. Share full context, not summaries — give each agent the relevant traces/facts, not a lossy digest.

This is orchestrator discipline applied before dispatch; do not ask subagents to coordinate with each other.

2. Create Focused Agent Tasks

Each agent gets:

• Specific scope: One test file or subsystem
• Clear goal: Make these tests pass
• Constraints: Don't change other code
• Expected output: Summary of what you found and fixed

3. Dispatch in Parallel

// In Claude Code / AI environment
Task("Fix agent-tool-abort.test.ts failures")
Task("Fix batch-completion-behavior.test.ts failures")
Task("Fix tool-approval-race-conditions.test.ts failures")
// All three run concurrently

4. Review and Integrate

When agents return:

• Read each summary
• Verify fixes don't conflict
• Run full test suite
• Integrate all changes

Use each worker's completion signal. Do not poll healthy workers. Record a phase budget for long work. A phase-overrun permits one status inspection and a revised decision. Do not use shell sleep with process, event, output, or status inspection.

Agent Prompt Structure

Good agent prompts are:

1. Focused - One clear problem domain
2. Self-contained - All context needed to understand the problem
3. Specific about output - What should the agent return?

Fix the 3 failing tests in src/agents/agent-tool-abort.test.ts:

1. "should abort tool with partial output capture" - expects 'interrupted at' in message
2. "should handle mixed completed and aborted tools" - fast tool aborted instead of completed
3. "should properly track pendingToolCount" - expects 3 results but gets 0

These are timing/race condition issues. Your task:

1. Read the test file and understand what each test verifies
2. Identify root cause - timing issues or actual bugs?
3. Fix by:
   - Replacing arbitrary timeouts with event-based waiting
   - Fixing bugs in abort implementation if found
   - Adjusting test expectations if testing changed behavior

Do NOT just increase timeouts - find the real issue.

Return: Summary of what you found and what you fixed.

Common Mistakes

❌ Too broad: "Fix all the tests" - agent gets lost ✅ Specific: "Fix agent-tool-abort.test.ts" - focused scope

❌ No context: "Fix the race condition" - agent doesn't know where ✅ Context: Paste the error messages and test names

❌ No constraints: Agent might refactor everything ✅ Constraints: "Do NOT change production code" or "Fix tests only"

❌ Vague output: "Fix it" - you don't know what changed ✅ Specific: "Return summary of root cause and changes"

When NOT to Use

Related failures: Fixing one might fix others - investigate together first Need full context: Understanding requires seeing entire system Exploratory debugging: You don't know what's broken yet Shared state: Agents would interfere (editing same files, using same resources) Single task: Only one task remaining — no parallelism benefit Same files: Tasks that modify the same files — merge conflicts likely even with worktree isolation

Integration

Invoked by:

• subagent-driven-development — uses this skill's generic concurrent-dispatch pattern after its scheduler proves graph work safe.
• getting-up-to-speed — heavy path (150+ tracked files) dispatches @researcher + @explore in parallel via this pattern.

Invokes: None — this is a dispatch pattern skill, not a pipeline skill.

SDD Integration

This skill does not decide whether graph tasks are independent. Route graph execution to superbeads:subagent-driven-development, whose sole scheduling owner is:

python3 "$PWD/skills/subagent-driven-development/scripts/sdd-scheduler.py" decide STATE.json

SDD validates dependencies, reviewed commits, contracts, write sets, exclusive/capacity resources, review reserve, host capability, and safe speculation before it invokes this generic pattern. Keep lifecycle, worktree, review, merge, and completion tutorials in SDD; do not duplicate them here.

Real Example from Session

Scenario: 6 test failures across 3 files after major refactoring

Failures:

• agent-tool-abort.test.ts: 3 failures (timing issues)
• batch-completion-behavior.test.ts: 2 failures (tools not executing)
• tool-approval-race-conditions.test.ts: 1 failure (execution count = 0)

Decision: Independent domains - abort logic separate from batch completion separate from race conditions

Dispatch:

Agent 1 → Fix agent-tool-abort.test.ts
Agent 2 → Fix batch-completion-behavior.test.ts
Agent 3 → Fix tool-approval-race-conditions.test.ts

Results:

• Agent 1: Replaced timeouts with event-based waiting
• Agent 2: Fixed event structure bug (threadId in wrong place)
• Agent 3: Added wait for async tool execution to complete

Integration: All fixes independent, no conflicts, full suite green

Time saved: 3 problems solved in parallel vs sequentially

Key Benefits

1. Parallelization - Multiple investigations happen simultaneously
2. Focus - Each agent has narrow scope, less context to track
3. Independence - Agents don't interfere with each other
4. Speed - 3 problems solved in time of 1

Verification

After agents return:

1. Review each summary - Understand what changed
2. Check for conflicts - Did agents edit same code?
3. Run full suite - Verify all fixes work together
4. Spot check - Agents can make systematic errors
5. No weakening to "pass" - An agent may not satisfy its narrow goal by weakening tests, dropping a requirement, or regressing security — verify this on integration (Production-Grade Doctrine)

When this workflow closes after producing durable insights, read Durable Memory and apply it; otherwise do not load it.

Real-World Impact

From debugging session (2025-10-03):

• 6 failures across 3 files
• 3 agents dispatched in parallel
• All investigations completed concurrently
• All fixes integrated successfully
• Zero conflicts between agent changes