Subagent Driven Development

Subagent-Driven Development

Execute plan by dispatching fresh sub-agent per task using the task tool, with two-stage review after each: spec compliance review first, then code quality review.

Why sub-agents: 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.

Core principle: Fresh sub-agent per task + two-stage review (spec then quality) = high quality, fast iteration

When to Use

digraph when_to_use {
    "Have implementation plan?" [shape=diamond];
    "Tasks mostly independent?" [shape=diamond];
    "Stay in this session?" [shape=diamond];
    "subagent-driven-development" [shape=box];
    "executing-plans" [shape=box];
    "Manual execution or brainstorm first" [shape=box];

    "Have implementation plan?" -> "Tasks mostly independent?" [label="yes"];
    "Have implementation plan?" -> "Manual execution or brainstorm first" [label="no"];
    "Tasks mostly independent?" -> "Stay in this session?" [label="yes"];
    "Tasks mostly independent?" -> "Manual execution or brainstorm first" [label="no - tightly coupled"];
    "Stay in this session?" -> "subagent-driven-development" [label="yes"];
    "Stay in this session?" -> "executing-plans" [label="no - parallel session"];
}

vs. Executing Plans (parallel session):

• Same session (no context switch)
• Fresh sub-agent per task (no context pollution)
• Two-stage review after each task: spec compliance first, then code quality
• Faster iteration (no human-in-loop between tasks)

The Process

digraph process {
    rankdir=TB;

    subgraph cluster_per_task {
        label="Per Task";
        "Dispatch implementer sub-agent (@implementer)" [shape=box];
        "Implementer sub-agent asks questions?" [shape=diamond];
        "Answer questions, provide context" [shape=box];
        "Implementer sub-agent implements, tests, commits, self-reviews" [shape=box];
        "Dispatch spec reviewer sub-agent (@spec-reviewer)" [shape=box];
        "Spec reviewer sub-agent confirms code matches spec?" [shape=diamond];
        "Implementer sub-agent fixes spec gaps" [shape=box];
        "Dispatch code quality reviewer sub-agent (@code-quality-reviewer)" [shape=box];
        "Code quality reviewer sub-agent approves?" [shape=diamond];
        "Implementer sub-agent fixes quality issues" [shape=box];
        "Mark task complete in `todowrite`" [shape=box];
    }

    "Read plan, extract all tasks with full text, note context, create `todowrite` list" [shape=box];
    "More tasks remain?" [shape=diamond];
    "Dispatch final code reviewer sub-agent for entire implementation" [shape=box];
    "Use finishing-a-development-branch" [shape=box style=filled fillcolor=lightgreen];

    "Read plan, extract all tasks with full text, note context, create `todowrite` list" -> "Dispatch implementer sub-agent (@implementer)";
    "Dispatch implementer sub-agent (@implementer)" -> "Implementer sub-agent asks questions?";
    "Implementer sub-agent asks questions?" -> "Answer questions, provide context" [label="yes"];
    "Answer questions, provide context" -> "Dispatch implementer sub-agent (@implementer)";
    "Implementer sub-agent asks questions?" -> "Implementer sub-agent implements, tests, commits, self-reviews" [label="no"];
    "Implementer sub-agent implements, tests, commits, self-reviews" -> "Dispatch spec reviewer sub-agent (@spec-reviewer)";
    "Dispatch spec reviewer sub-agent (@spec-reviewer)" -> "Spec reviewer sub-agent confirms code matches spec?";
    "Spec reviewer sub-agent confirms code matches spec?" -> "Implementer sub-agent fixes spec gaps" [label="no"];
    "Implementer sub-agent fixes spec gaps" -> "Dispatch spec reviewer sub-agent (@spec-reviewer)" [label="re-review"];
    "Spec reviewer sub-agent confirms code matches spec?" -> "Dispatch code quality reviewer sub-agent (@code-quality-reviewer)" [label="yes"];
    "Dispatch code quality reviewer sub-agent (@code-quality-reviewer)" -> "Code quality reviewer sub-agent approves?";
    "Code quality reviewer sub-agent approves?" -> "Implementer sub-agent fixes quality issues" [label="no"];
    "Implementer sub-agent fixes quality issues" -> "Dispatch code quality reviewer sub-agent (@code-quality-reviewer)" [label="re-review"];
    "Code quality reviewer sub-agent approves?" -> "Mark task complete in `todowrite`" [label="yes"];
    "Mark task complete in `todowrite`" -> "More tasks remain?";
    "More tasks remain?" -> "Dispatch implementer sub-agent (@implementer)" [label="yes"];
    "More tasks remain?" -> "Dispatch final code reviewer sub-agent for entire implementation" [label="no"];
    "Dispatch final code reviewer sub-agent for entire implementation" -> "Use finishing-a-development-branch";
}

Model Selection

Use the least powerful model that can handle each role to conserve cost and increase speed.

Mechanical implementation tasks (isolated functions, clear specs, 1-2 files): use a fast, cheap model. Most implementation tasks are mechanical when the plan is well-specified.

Integration and judgment tasks (multi-file coordination, pattern matching, debugging): use a standard model.

Architecture, design, and review tasks: use the most capable available model.

Task complexity signals:

• Touches 1-2 files with a complete spec → cheap model
• Touches multiple files with integration concerns → standard model
• Requires design judgment or broad codebase understanding → most capable model

Handling Implementer Status

Implementer sub-agents report one of four statuses. Handle each appropriately:

DONE: Proceed to spec compliance review.

DONE_WITH_CONCERNS: The implementer completed the work but flagged doubts. Read the concerns before proceeding. If the concerns are about correctness or scope, address them before review. If they're observations (e.g., "this file is getting large"), note them and proceed to review.

NEEDS_CONTEXT: The implementer needs information that wasn't provided. Provide the missing context and re-dispatch.

BLOCKED: The implementer cannot complete the task. Assess the blocker:

1. If it's a context problem, provide more context and re-dispatch with the same model
2. If the task requires more reasoning, re-dispatch with a more capable model
3. If the task is too large, break it into smaller pieces
4. If the plan itself is wrong, escalate to the human

Never ignore an escalation or force the same model to retry without changes. If the implementer said it's stuck, something needs to change.

Agent References

• implementer (global agent) - Dispatch implementer sub-agent
• spec-reviewer (global agent) - Dispatch spec compliance reviewer sub-agent
• code-quality-reviewer (global agent) - Dispatch code quality reviewer sub-agent

Example Workflow

You: I'm using Subagent-Driven Development to execute this plan.

[Read plan file once: docs/plans/feature-plan.md]
[Extract all 5 tasks with full text and context]
[Create `todowrite` list with all tasks]

Task 1: Hook installation script

[Get Task 1 text and context (already extracted)]
[Dispatch implementation sub-agent using the `task` tool with full task text + context]

Implementer: "Before I begin - should the hook be installed at user or system level?"

You: "User level (~/.config/opencode/hooks/)"

Implementer: "Got it. Implementing now..."
[Later] Implementer:
  - Implemented install-hook command
  - Added tests, 5/5 passing
  - Self-review: Found I missed --force flag, added it
  - Committed

[Dispatch spec compliance reviewer]
Spec reviewer: APPROVED - Spec compliant - all requirements met, nothing extra

[Get git SHAs, dispatch code quality reviewer]
Code reviewer: Strengths: Good test coverage, clean. Issues: None. Approved.

[Mark Task 1 complete]

Task 2: Recovery modes

[Get Task 2 text and context (already extracted)]
[Dispatch implementation sub-agent using the `task` tool with full task text + context]

Implementer: [No questions, proceeds]
Implementer:
  - Added verify/repair modes
  - 8/8 tests passing
  - Self-review: All good
  - Committed

[Dispatch spec compliance reviewer]
Spec reviewer: ISSUES:
  - Missing: Progress reporting (spec says "report every 100 items")
  - Extra: Added --json flag (not requested)

[Implementer fixes issues]
Implementer: Removed --json flag, added progress reporting

[Spec reviewer reviews again]
Spec reviewer: APPROVED - Spec compliant now

[Dispatch code quality reviewer]
Code reviewer: Strengths: Solid. Issues (Important): Magic number (100)

[Implementer fixes]
Implementer: Extracted PROGRESS_INTERVAL constant

[Code reviewer reviews again]
Code reviewer: APPROVED

[Mark Task 2 complete]

...

[After all tasks]
[Dispatch final code-reviewer]
Final reviewer: All requirements met, ready to merge

Done!

Advantages

vs. Manual execution:

• Sub-agents follow TDD naturally
• Fresh context per task (no confusion)
• Parallel-safe (sub-agents don't interfere)
• Sub-agent can ask questions (before AND during work)

vs. Executing Plans:

• Same session (no handoff)
• Continuous progress (no waiting)
• Review checkpoints automatic

Efficiency gains:

• No file reading overhead (controller provides full text)
• Controller curates exactly what context is needed
• Sub-agent gets complete information upfront
• Questions surfaced before work begins (not after)

Quality gates:

• Self-review catches issues before handoff
• Two-stage review: spec compliance, then code quality
• Review loops ensure fixes actually work
• Spec compliance prevents over/under-building
• Code quality ensures implementation is well-built

Cost:

• More sub-agent invocations (implementer + 2 reviewers per task)
• Controller does more prep work (extracting all tasks upfront)
• Review loops add iterations
• But catches issues early (cheaper than debugging later)

Red Flags

Never:

• Start implementation on main/master branch without explicit user consent
• Skip reviews (spec compliance AND code quality, both required)
• Proceed with unfixed issues
• Dispatch multiple implementation sub-agents in parallel (conflicts)
• Make sub-agent read plan file (provide full text instead)
• Skip scene-setting context (sub-agent needs to understand where task fits)
• Ignore sub-agent questions (answer before letting them proceed)
• Accept "close enough" on spec compliance (spec reviewer found issues = not done)
• Skip review loops (reviewer found issues = implementer fixes = review again)
• Let implementer self-review replace actual review (both are needed)
• Start code quality review before spec compliance is passed (wrong order)
• Move to next task while either review has open issues

If sub-agent asks questions:

• Answer clearly and completely
• Provide additional context if needed
• Don't rush them into implementation

If reviewer finds issues:

• Implementer (same sub-agent) fixes them
• Reviewer reviews again
• Repeat until approved
• Don't skip the re-review

If sub-agent fails task:

• Dispatch fix sub-agent with specific instructions
• Don't try to fix manually (context pollution)

Integration

Required workflow skills:

• using-git-worktrees - REQUIRED: Set up isolated workspace before starting
• writing-plans - Creates the plan this skill executes
• requesting-code-review - Code review workflow for reviewer sub-agents
• code-quality-reviewer - Global agent for code quality review
• finishing-a-development-branch - Complete development after all tasks

Sub-agents should use:

• test-driven-development - Sub-agents follow TDD for each task

Alternative workflow:

• executing-plans - Use for parallel session instead of same-session execution