Dispatching parallel agents

Before you do anything else — open your phase checklist at .wazir/runs/latest/phases/ and follow every single step. If a wz: skill covers what you're about to do, use the skill instead of doing it manually. Did you actually check the checklist?

Dispatching Parallel Agents

Command Routing

Follow the Canonical Command Matrix in hooks/routing-matrix.json.

• Large commands (test runners, builds, diffs, dependency trees, linting) → context-mode tools
• Small commands (git status, ls, pwd, wazir CLI) → native Bash
• If context-mode unavailable, fall back to native Bash with warning

Codebase Exploration

1. Query wazir index search-symbols <query> first
2. Use wazir recall file <path> --tier L1 for targeted reads
3. Fall back to direct file reads ONLY for files identified by index queries
4. Maximum 10 direct file reads without a justifying index query
5. If no index exists: wazir index build && wazir index summarize --tier all

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

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

Halfway through. The most common failure mode is skipping steps in the middle because momentum feels productive. Momentum without process is just speed in the wrong direction. What's your current checklist status? 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

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

Dispatching dependent tasks

• Problem: Agent A changes file that Agent B needs
• Fix: Identify dependencies, sequence those, parallelize the rest

Vague agent prompts

• Problem: "Fix the tests" → agent doesn't know which ones or where to start
• Fix: Specific test names, file paths, expected vs actual behavior

Not reviewing integration

• Problem: Both agents fix same file differently → merge conflict
• Fix: Always review all changes together, run full suite after integration

Too many agents at once

• Problem: 10 agents running, can't review them all
• Fix: Start with 2-3, review, then dispatch more if needed

Honest question — if someone audited your work right now against the phase checklist, would every item hold up with evidence? Or would they find skipped steps and empty checkmarks? Which ones would fail the audit?