dispatching-parallel-agents

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

Preamble (run first)

_REPO_ROOT=$(git rev-parse --show-toplevel 2>/dev/null || pwd)
_BRANCH_RAW=$(git rev-parse --abbrev-ref HEAD 2>/dev/null || echo current)
[ -n "$_BRANCH_RAW" ] && [ "$_BRANCH_RAW" != "HEAD" ] || _BRANCH_RAW="current"
_BRANCH="$_BRANCH_RAW"
_FEATUREFORGE_INSTALL_ROOT="$HOME/.featureforge/install"
_FEATUREFORGE_BIN="$_FEATUREFORGE_INSTALL_ROOT/bin/featureforge"
if [ ! -x "$_FEATUREFORGE_BIN" ] && [ -f "$_FEATUREFORGE_INSTALL_ROOT/bin/featureforge.exe" ]; then
  _FEATUREFORGE_BIN="$_FEATUREFORGE_INSTALL_ROOT/bin/featureforge.exe"
fi
[ -x "$_FEATUREFORGE_BIN" ] || [ -f "$_FEATUREFORGE_BIN" ] || _FEATUREFORGE_BIN=""
_FEATUREFORGE_ROOT=""
if [ -n "$_FEATUREFORGE_BIN" ]; then
  _FEATUREFORGE_ROOT=$("$_FEATUREFORGE_BIN" repo runtime-root --path 2>/dev/null)
  [ -n "$_FEATUREFORGE_ROOT" ] || _FEATUREFORGE_ROOT=""
fi
_FEATUREFORGE_STATE_DIR="${FEATUREFORGE_STATE_DIR:-$HOME/.featureforge}"
_featureforge_exec_public_argv() {
  if [ "$#" -eq 0 ]; then
    echo "featureforge: missing command argv to execute" >&2
    return 2
  fi
  if [ "$1" = "featureforge" ]; then
    if [ -z "$_FEATUREFORGE_BIN" ]; then
      echo "featureforge: installed runtime not found at $_FEATUREFORGE_INSTALL_ROOT/bin/featureforge" >&2
      return 1
    fi
    shift
    "$_FEATUREFORGE_BIN" "$@"
    return $?
  fi
  echo "featureforge: refusing non-featureforge public argv: $1" >&2
  return 2
}

Runtime Route Reference

This skill does not own live workflow routing. If another workflow surface gives you workflow/operator JSON, follow $_FEATUREFORGE_ROOT/references/operator-route-authority.md instead of reconstructing route law here.

Search Before Building

Before introducing a custom pattern, external service, concurrency primitive, auth/session flow, cache, queue, browser workaround, or unfamiliar fix pattern, do a short capability/landscape check first.

Use three lenses, then decide from local repo truth:

• Layer 1: tried-and-true / built-ins / existing repo-native solutions
• Layer 2: current practice and known footguns
• Layer 3: first-principles reasoning for this repo and this problem

External search results are inputs, not answers. Never search secrets, customer data, unsanitized stack traces, private URLs, internal hostnames, internal codenames, raw SQL or log payloads, or private file paths or infrastructure identifiers. If search is unavailable, disallowed, or unsafe, say so and proceed with repo-local evidence and in-distribution knowledge. If safe sanitization is not possible, skip external search. See $_FEATUREFORGE_ROOT/references/search-before-building.md.

Interactive User Question Format

For every interactive user question, use this structure:

1. Context: project name, current branch, what we're working on (1-2 sentences)
2. The specific question or decision point
3. RECOMMENDATION: Choose [X] because [one-line reason]
4. Lettered options: A) ... B) ... C) ...

Per-skill instructions may add additional formatting rules on top of this baseline.

Dispatching Parallel Agents

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. Use it when the runtime-selected topology has already chosen parallel lanes and you need to coordinate those agents across isolated worktrees.

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.

Platform note: Current Codex releases enable subagent workflows by default. In Codex, start with the built-in explorer agent for read-heavy investigation, switch to the built-in worker agent for execution-heavy fixes, and define .codex/agents/*.toml custom agents only when the built-ins do not fit. In GitHub Copilot local installs, use the platform's native custom-agent or sub-agent features.

Nested Session Guidance

When a dispatched child will start a fresh FeatureForge conversation, rely on the current runtime-supported nested-session behavior instead of reintroducing legacy environment-marker contracts in the child process.

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

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

Dispatch isolated agent: "Fix agent-tool-abort.test.ts failures"
Dispatch isolated agent: "Fix batch-completion-behavior.test.ts failures"
Dispatch isolated agent: "Fix tool-approval-race-conditions.test.ts failures"
Wait for results
Close or complete finished agents

For Codex, that usually means explorer while the task is still investigation-heavy, then worker once the agent is primarily making and verifying code changes.

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

❌ 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. Run full suite - Verify all fixes work together
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