Subagent-Driven Development

Use when executing implementation plans with independent tasks in the current session

Security Scan

The skill's instructions are coherent with a developer orchestration workflow, but it omits declaring needed runtime privileges (filesystem/git access, connectors like TodoWrite/superpowers) and contains minor instruction inconsistencies — proceed with caution and verify integrations and permissions before use.

Purpose & Capability

Name/description match the instructions: the skill orchestrates fresh subagents per task with spec + quality reviews. However, the runtime behavior implies read/write access to the repository (read plans, read/write code, run tests, commit, obtain git SHAs) and potentially external integrations (TodoWrite, 'superpowers' tools) but the skill declares no required binaries, env vars, or config paths. The lack of stated required permissions is a design omission that should be clarified.

Instruction Scope

SKILL.md instructs agents to read plan files and project files, implement code, run tests, commit, inspect code line-by-line, and mark tasks complete in TodoWrite — all within-scope for a dev orchestration skill. However: (1) prompts explicitly tell implementer subagents to paste 'FULL TEXT' instead of reading files while examples show reading a plan file — inconsistent guidance; (2) the workflow grants broad access to repository contents and commit history and to any context provided to subagents, which can expose secrets or unrelated sensitive files if not confined; (3) the skill references external tools (superpowers:code-reviewer, TodoWrite, etc.) without describing where those endpoints live or what credentials they need.

Install Mechanism

Instruction-only skill with no install spec and no code files. This is low-risk from an installation/code distribution perspective because nothing is downloaded or written by an installer.

Credentials

The skill declares no required environment variables or credentials, yet its intended actions imply the need for: repository write access (git), possibly CI/test runners, and credentials for external services (TodoWrite, 'superpowers' tools). The omission means callers might need to grant broad runtime permissions implicitly; any unspecified external connectors or tokens are a risk because they could be used to exfiltrate data or make changes without explicit declaration.

Persistence & Privilege

always:false and user-invocable:true — normal and appropriate. The skill does instruct autonomous dispatch of subagents, but autonomous invocation is platform-default. There is no evidence the skill attempts to modify other skills or system-wide settings.

What to consider before installing

This skill is an orchestration template for spawning implementer/spec/quality-review subagents and expects access to your repo and developer tools, but it doesn't explicitly declare those permissions. Before installing or using it: 1) Confirm what 'superpowers' and 'TodoWrite' integrations are and whether they require tokens or network endpoints; don't grant unknown tokens. 2) Run it first in an isolated/test workspace (or sandbox) so subagents only see non-sensitive files. 3) Ensure the agent runtime only has the minimum filesystem and git permissions needed (least privilege) and that commits/actions are audited. 4) Review the prompt templates to ensure they don't instruct subagents to leak session context or secrets (search for commands that send data to external URLs). 5) Ask the publisher for clarification about required permissions and expected external connectors; if they can't justify missing env/config declarations, treat the skill with extra caution.

Execute plan by dispatching fresh subagent per task, with two-stage review after each: spec compliance review first, then code quality review.

Core principle: Fresh subagent 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 subagent 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 subagent (./implementer-prompt.md)" [shape=box];
        "Implementer subagent asks questions?" [shape=diamond];
        "Answer questions, provide context" [shape=box];
        "Implementer subagent implements, tests, commits, self-reviews" [shape=box];
        "Dispatch spec reviewer subagent (./spec-reviewer-prompt.md)" [shape=box];
        "Spec reviewer subagent confirms code matches spec?" [shape=diamond];
        "Implementer subagent fixes spec gaps" [shape=box];
        "Dispatch code quality reviewer subagent (./code-quality-reviewer-prompt.md)" [shape=box];
        "Code quality reviewer subagent approves?" [shape=diamond];
        "Implementer subagent fixes quality issues" [shape=box];
        "Mark task complete in TodoWrite" [shape=box];
    }

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

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

Prompt Templates

• ./implementer-prompt.md - Dispatch implementer subagent
• ./spec-reviewer-prompt.md - Dispatch spec compliance reviewer subagent
• ./code-quality-reviewer-prompt.md - Dispatch code quality reviewer subagent

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 with all tasks]

Task 1: Hook installation script

[Get Task 1 text and context (already extracted)]
[Dispatch implementation subagent with full task text + context]

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

You: "User level (~/.config/superpowers/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: ✅ 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 subagent 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: ✅ 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:

• Subagents follow TDD naturally
• Fresh context per task (no confusion)
• Parallel-safe (subagents don't interfere)
• Subagent 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
• Subagent 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 subagent 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:

• Skip reviews (spec compliance OR code quality)
• Proceed with unfixed issues
• Dispatch multiple implementation subagents in parallel (conflicts)
• Make subagent read plan file (provide full text instead)
• Skip scene-setting context (subagent needs to understand where task fits)
• Ignore subagent 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 ✅ (wrong order)
• Move to next task while either review has open issues

If subagent asks questions:

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

If reviewer finds issues:

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

If subagent fails task:

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

Integration

Required workflow skills:

• superpowers:writing-plans - Creates the plan this skill executes
• superpowers:requesting-code-review - Code review template for reviewer subagents
• superpowers:finishing-a-development-branch - Complete development after all tasks

Subagents should use:

• superpowers:test-driven-development - Subagents follow TDD for each task

Alternative workflow:

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

Handling Subagent Help Requests

When subagent sends <request_help>:

Identify Priority & Type

Priority	Response Time	Action
🔴 High	Immediate	Stop everything, address first
🟡 Medium	Within task	Address before proceeding
🟢 Low	End of task	Note it, address later

Respond by Type

ambiguity → Clarify the task description

<help_response>
  <clarification>具体意思是 X，不是 Y</clarification>
  <example>比如这个场景下应该...</example>
</help_response>

blocked → Remove the blocker or provide workaround

<help_response>
  <solution>我已经创建了这个依赖文件</solution>
  <workaround>或者你可以先用这个临时方案...</workaround>
</help_response>

conflict → Make the call

<help_response>
  <decision>用方案 A</decision>
  <reason>因为 B 会导致 C 问题</reason>
</help_response>

decision → Make the decision and explain

<help_response>
  <choice>选 X 方案</choice>
  <why>因为 Y，而且 Z</why>
  <tradeoffs>代价是 A，但可以接受因为 B</tradeoffs>
</help_response>

scope → Either expand scope or handle it yourself

<help_response>
  <scope_decision>expanded | separate_task | lead_handles</scope_decision>
  <instruction>如果是 expanded：这个也算你的任务 / 如果 separate_task：记下来稍后处理 / 如果 leadHandles：我来搞</instruction>
</help_response>

critical → Take over or guide through recovery

<help_response>
  <action>我正在介入处理</action>
  <guidance>你先做 X，Y 留给我</guidance>
</help_response>

Response Quality Checklist

• ✅ Clear and specific (no vague answers)
• ✅ Actionable (subagent knows exactly what to do next)
• ✅ Explains why (so subagent learns)
• ✅ Addresses the priority appropriately
• ❌ No "figure it out yourself"
• ❌ No skipping the response