tree-of-thoughts

Key benefits:

• Systematic exploration - Multiple agents explore different regions of the solution space
• Structured evaluation - Meta-judges produce tailored rubrics and criteria before judging
• Independent verification - Judges apply meta-judge specifications mechanically, reducing bias
• Adaptive strategy - Clear winners get polished, split decisions get synthesized, failures get redesigned

Pattern: Tree of Thoughts (ToT)

This command implements an eight-phase systematic reasoning pattern with meta-judge evaluation and adaptive strategy selection:

Phase 1: Exploration (Propose Approaches)
         ┌─ Agent A → Proposals A1, A2 (with probabilities) ─┐
Task ───┼─ Agent B → Proposals B1, B2 (with probabilities) ─┼─┐
         └─ Agent C → Proposals C1, C2 (with probabilities) ─┘ │
                                                                │
Phase 1.5: Pruning Meta-Judge (runs in parallel with Phase 1) │
         Meta-Judge → Pruning Evaluation Specification YAML ───┤
                                                                │
Phase 2: Pruning (Vote for Best 3)                             │
         ┌─ Judge 1 → Votes + Rationale ─┐                     │
         ├─ Judge 2 → Votes + Rationale ─┼─────────────────────┤
         └─ Judge 3 → Votes + Rationale ─┘                     │
                 │                                              │
                 ├─→ Select Top 3 Proposals                     │
                 │                                              │
Phase 3: Expansion (Develop Full Solutions)                    │
         ┌─ Agent A → Solution A (from proposal X) ─┐          │
         ├─ Agent B → Solution B (from proposal Y) ─┼──────────┤
         └─ Agent C → Solution C (from proposal Z) ─┘          │
                                                                │
Phase 3.5: Evaluation Meta-Judge (runs in parallel w/ Phase 3)│
         Meta-Judge → Evaluation Specification YAML ───────────┤
                                                                │
Phase 4: Evaluation (Judge Full Solutions)                     │
         ┌─ Judge 1 → Report 1 ─┐                              │
         ├─ Judge 2 → Report 2 ─┼──────────────────────────────┤
         └─ Judge 3 → Report 3 ─┘                              │
                                                                │
Phase 4.5: Adaptive Strategy Selection                         │
         Analyze Consensus ────────────────────────────────────┤
                ├─ Clear Winner? → SELECT_AND_POLISH           │
                ├─ All Flawed (<3.0)? → REDESIGN (Phase 3)     │
                └─ Split Decision? → FULL_SYNTHESIS            │
                                         │                      │
Phase 5: Synthesis (Only if FULL_SYNTHESIS)                    │
         Synthesizer ────────────────────┴──────────────────────┴─→ Final Solution

Process

Setup: Create Directory Structure

Before starting, ensure the directory structure exists:

mkdir -p .specs/research .specs/reports

Naming conventions:

• Proposals: .specs/research/{solution-name}-{YYYY-MM-DD}.proposals.[a|b|c].md
• Pruning: .specs/research/{solution-name}-{YYYY-MM-DD}.pruning.[1|2|3].md
• Selection: .specs/research/{solution-name}-{YYYY-MM-DD}.selection.md
• Evaluation: .specs/reports/{solution-name}-{YYYY-MM-DD}.[1|2|3].md

Where:

• {solution-name} - Derived from output path (e.g., users-api from output specs/api/users.md)
• {YYYY-MM-DD} - Current date

Note: Solutions remain in their specified output locations; only research and evaluation files go to .specs/

Phase 1: Exploration (Propose Approaches)

Launch 3 independent agents in parallel (recommended: Sonnet for speed):

1. Each agent receives identical task description and context
2. Each agent generates 6 high-level approaches (not full implementations)
3. For each approach, agent provides:
   • Approach description (2-3 paragraphs)
   • Key design decisions and trade-offs
   • Probability estimate (0.0-1.0)
   • Estimated complexity (low/medium/high)
   • Potential risks and failure modes
4. Proposals saved to .specs/research/{solution-name}-{date}.proposals.[a|b|c].md

Key principle: Systematic exploration through probabilistic sampling from the full distribution of possible approaches.

Prompt template for explorers:

<task>
{task_description}
</task>

<constraints>
{constraints_if_any}
</constraints>

<context>
{relevant_context}
</context>

<output>
{.specs/research/{solution-name}-{date}.proposals.[a|b|c].md - each agent gets unique letter identifier}
</output>

Instructions:

Let's approach this systematically by first understanding what we're solving, then exploring the solution space.

**Step 1: Decompose the problem**
Before generating approaches, break down the task:
- What is the core problem being solved?
- What are the key constraints and requirements?
- What subproblems must any solution address?
- What are the evaluation criteria for success?

**Step 2: Map the solution space**
Identify the major dimensions along which solutions can vary:
- Architecture patterns (e.g., monolithic vs distributed)
- Implementation strategies (e.g., eager vs lazy)
- Trade-off axes (e.g., performance vs simplicity)

**Step 3: Generate 6 distinct high-level approaches**

**Sampling guidance:**
Please sample approaches at random from the [full distribution / tails of the distribution]
- For first 3 approaches aim for high probability, over 0.80
- For last 3 approaches aim for diversity - explore different regions of the solution space, such that the probability of each response is less than 0.10

For each approach, provide:
   - Name and one-sentence summary
   - Detailed description (2-3 paragraphs)
   - Key design decisions and rationale
   - Trade-offs (what you gain vs what you sacrifice)
   - Probability (0.0-1.0)
   - Complexity estimate (low/medium/high)
   - Potential risks and failure modes

**Step 4: Verify diversity**
Before finalizing, check:
- Are approaches genuinely different, not minor variations?
- Do they span different regions of the solution space?
- Have you covered both conventional and unconventional options?


CRITICAL:
- Do NOT implement full solutions yet - only high-level approaches
- Ensure approaches are genuinely different, not minor variations

Phase 1.5: Dispatch Pruning Meta-Judge

CRITICAL: Launch the pruning meta-judge in parallel with Phase 1 exploration agents. The meta-judge does not need exploration output to generate pruning criteria — it only needs the original task description.

The pruning meta-judge generates an evaluation specification (rubrics, checklist, scoring criteria) tailored to evaluating high-level proposals for pruning.

Prompt template for pruning meta-judge:

## Task

Generate an evaluation specification yaml for pruning high-level solution proposals. You will produce rubrics, checklists, and scoring criteria that judge agents will use to select the top 3 proposals for full development.

CLAUDE_PLUGIN_ROOT=`${CLAUDE_PLUGIN_ROOT}`

## User Prompt
{Original task description from user}

## Context
{Any relevant codebase context, file paths, constraints}

## Artifact Type
proposals (high-level approaches with probability estimates, not full implementations)

## Evaluation Focus
Feasibility, alignment with requirements, potential for high-quality result, risk manageability

## Instructions
Return only the final evaluation specification YAML in your response.
The specification should support comparative evaluation and ranking of proposals.

Dispatch:

Use Task tool:
  - description: "Pruning Meta-judge: {brief task summary}"
  - prompt: {pruning meta-judge prompt}
  - model: opus
  - subagent_type: "sadd:meta-judge"

Phase 2: Pruning (Vote for Top 3 Candidates)

Wait for BOTH Phase 1 exploration agents AND Phase 1.5 pruning meta-judge to complete before proceeding.

Launch 3 independent judges in parallel (recommended: Opus for rigor):

1. Each judge receives ALL proposal files (from .specs/research/) and the pruning meta-judge evaluation specification YAML
2. Judges evaluate each proposal against the meta-judge-generated pruning criteria
3. Each judge produces:
   • Scores for each proposal (with evidence)
   • Vote for top 3 proposals to expand
   • Rationale for selections
4. Votes saved to .specs/research/{solution-name}-{date}.pruning.[1|2|3].md

Key principle: Independent evaluation with meta-judge-generated criteria ensures consistent, tailored assessment without hardcoded weights.

CRITICAL: Provide to each judge the EXACT pruning meta-judge's evaluation specification YAML. Do not skip, add, modify, shorten, or summarize any text in it!

Prompt template for pruning judges:

You are evaluating {N} proposed approaches against an evaluation specification produced by the meta judge, to select the top 3 for full development.

CLAUDE_PLUGIN_ROOT=`${CLAUDE_PLUGIN_ROOT}`

## Task
{task_description}

## Proposals
{list of paths to all proposal files}
Read all proposals carefully before evaluating.

## Evaluation Specification

```yaml
{pruning meta-judge's evaluation specification YAML}

Output

{.specs/research/{solution-name}-{date}.pruning.[1|2|3].md}

Instructions

Follow your full judge process as defined in your agent instructions!

CRITICAL: You must reply with this exact structured evaluation report format in YAML at the START of your response!

**Dispatch:**

Use Task tool:

• description: "Pruning Judge {1|2|3}: {brief task summary}"
• prompt: {pruning judge prompt with exact meta-judge specification YAML}
• model: opus
• subagent_type: "sadd:judge"

### Phase 2b: Select Top 3 Proposals

After judges complete voting:

1. **Aggregate votes** using ranked choice:
   - 1st choice = 3 points
   - 2nd choice = 2 points
   - 3rd choice = 1 point
2. **Select top 3** proposals by total points
3. **Handle ties** by comparing average scores across criteria
4. **Document selection** in `.specs/research/{solution-name}-{date}.selection.md`:
   - Vote tallies
   - Selected proposals
   - Consensus rationale

### Phase 3: Expansion (Develop Full Solutions)

Launch **3 independent agents in parallel** (recommended: Opus for quality):

1. Each agent receives:
   - **One selected proposal** to expand
   - **Original task description** and context
   - **Judge feedback** from pruning phase (concerns, questions)
2. Agent produces **complete solution** implementing the proposal:
   - Full implementation details
   - Addresses concerns raised by judges
   - Documents key decisions made during expansion
3. Solutions saved to `solution.a.md`, `solution.b.md`, `solution.c.md`

**Key principle:** Focused development of validated approaches with awareness of evaluation feedback.

**Prompt template for expansion agents:**

```markdown
You are developing a full solution based on a selected proposal.

<task>
{task_description}
</task>

<selected_proposal>
{write selected proposal EXACTLY as it is. Including all details provided by the agent}
Read this carefully - it is your starting point.
</selected_proposal>

<judge_feedback>
{concerns and questions from judges about this proposal}
Address these in your implementation.
</judge_feedback>

<output>
solution.[*].md where [*] is your unique identifier (a, b, or c)
</output>

Instructions:

Let's work through this systematically to ensure we build a complete, high-quality solution.

**Step 1: Understand the proposal deeply**
Before implementing, analyze:
- What is the core insight or approach of this proposal?
- What are the key design decisions already made?
- What gaps need to be filled for a complete solution?

**Step 2: Address judge feedback**
For each concern raised by judges:
- What specific change or addition addresses this concern?
- How does this change integrate with the proposal's approach?

**Step 3: Decompose int