dispatching-parallel-agents（调度并行 Agent）

Dispatching Parallel Agents

Dispatch one agent per independent problem. Let them work concurrently.

Dispatch Workflow

Copy and track:

- [ ] 1. Identify independent domains
- [ ] 2. Create focused agent tasks
- [ ] 3. Dispatch in parallel
- [ ] 4. Review and integrate

1. Identify Independent Domains

Group failures by what's broken:

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

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

Critical check: If fixing one might fix others → investigate together first (don't parallelize).

2. Create Focused Agent Tasks

Each agent needs:

• Scope: One test file or subsystem
• Goal: Make these tests pass
• Constraints: Don't change unrelated code
• Output: Summary of findings and fixes

3. Dispatch in Parallel

Example (Claude Code):

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")

4. Review and Integrate

1. Read each agent's summary
2. Check for conflicts (same files edited?)

- If two agents touched the same file → stop and re-scope (one owner per file)

1. Run full test suite
2. If failures:

- Check for merge conflicts → resolve manually - If no conflicts → investigate as new failures

1. Repeat until green

Agent Prompt Template

Fix the [N] failing tests in [file path]:

1. "[test name]" - [error summary]
2. "[test name]" - [error summary]

Context: [relevant background, e.g., "These are timing/race condition issues"]

Your task:
1. Read the test file, understand what each test verifies
2. Identify root cause—timing issues or actual bugs?
3. Fix by [preferred approach, e.g., "replacing arbitrary timeouts with event-based waiting"]

Do NOT: [anti-patterns, e.g., "just increase timeouts—find the real issue"]

Return: Summary of root cause and changes made.

Common Mistakes

Example

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

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