---
name: dispatching-parallel-agents
description: Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
---

# Dispatching Parallel Agents

Audience: AI coding agents first.

MUST follow `/home/user/Projects/0 DOCS/GIT_FATIGUE.md` §12. Per-agent work returns minimal task-test evidence; candidate-wide broad gate uses one valid receipt on assembled tree.

## 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 unrelated failures (different test files, different subsystems, different bugs), investigating them sequentially wastes time. Each investigation is independent and can happen in parallel.

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

## When to Use

```dot
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

**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.

### 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

```typescript
// 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
- Read assembled-tree receipt; absent/invalid receipt → designated executor runs full suite once
- Integrate all changes

## 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?

```markdown
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)

## 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. **Consume candidate receipt** - Missing/invalid receipt → designated executor runs full suite once after all fixes
4. **Spot check** - Agents can make systematic errors

## 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

## Learned Rules

### grep-comments-inflate-count | fired:1 | 2026-04-29
Used `grep -rln 'QueryClientProvider\|new QueryClient'` to count violations → 31 files. 10 were already migrated but mentioned "QueryClientProvider" in doc comments. Reported wrong scope to user and advisor.
Prevent: anchor grep to actual code — use `grep -rln '^import.*QueryClientProvider'` for imports, `grep -rln 'new QueryClient('` for instantiations. Never use bare keyword grep on file content to count violations; comments inflate count.

### bg-general-purpose-premature-exit | fired:4 | 2026-05-19
bg general-purpose subagents on impl/verify tasks → narration-as-summary after 2-3 tools, zero commits. Wave 2b retried 3× before switch to foreground `claude` type worked.
Prevent: impl/verify tasks with commit deliverable → always `subagent_type: "claude"`, foreground (no `run_in_background`). bg general-purpose = read-only research only. Before dispatching: "does this need a commit?" → claude type, foreground.

### impl-agent-spec-format-drift | fired:1 | 2026-05-19
Impl subagent read spec but produced own telemetry format (e.g. `prune ... N dropped (zero=X, low=Y)`) instead of spec-mandated exact strings. Required extra fix wave.
Prevent: copy-paste exact required output strings from spec into impl agent prompt verbatim. Never trust agent to infer format from spec reference alone — quote it word-for-word in the prompt.

### cf-token-in-deploy-subagent-brief | fired:1 | 2026-05-22
Deploy subagent hit Cloudflare MCP OAuth prompt → blocked, wasted ~437s + 64k tokens. Subagent can't open browser. Second dispatch with `CLOUDFLARE_API_TOKEN=<token>` in env worked immediately.
Prevent: NEVER dispatch deploy/wrangler publication to subagents. Submit verified candidate to delivery controller; never inline publication credentials in prompts.

### trace-imports-before-dispatch-not-after | fired:1 | 2026-05-22
First 503-fix subagent told "trace ALL static imports from middleware.ts, find heavy ones" — broad instruction. Subagent found only @sentry/cloudflare (obvious), missed session.ts→drizzle-orm chain (real root cause). Wasted full dispatch. Orchestrator then manually traced imports, found exact bleed paths, dispatched targeted second agent → success.
Prevent: orchestrator must trace import chain ITSELF (read 3-5 files) before dispatching fix agent. Provide exact file paths + exact import lines to change in the brief. Never delegate diagnosis + fix in one agent — diagnosis quality drops when combined with implementation pressure.

### post-compaction-edit-needs-reread | fired:1 | 2026-05-30
After context compaction, called Edit on `pages/purchases/[id]/index.astro` that had been read pre-compaction → "File has not been read yet. Read it first" error → wasted one round-trip re-reading.
Prevent: on compaction resume, treat ALL files as unread regardless of session summary. Before any Edit call, always Read the target file in the current context window first.