📦 @goodandready/dsh-moa

---

⚡ Overview & The Problem

Single-model AI generation often suffers from blind spots, single-perspective biases, hallucinated architectural choices, and inconsistent code quality on challenging engineering tasks. When prompted with ambiguous or complex specifications, a single model may make premature assumptions and produce monolithic, unvetted implementations.

@goodandready/dsh-moa brings the Mixture of Agents (MoA) architecture natively to DeepSeek Harness via the /moa slash command:

1. Adaptive Clarification Questionnaire: For broad or underspecified prompts, advisor models formulate clarifying options and the judge synthesizes a structured 2–4 question questionnaire before generating code.
2. Parallel Proposers Fan-Out & Workspace Isolation: Multiple independent models evaluate the prompt concurrently. Each candidate's proposed files are written to isolated disk sandboxes (.moa/candidate-N/), avoiding cross-pollution.
3. Frontier Judge Evaluation & File Promotion: A flagship reasoning model critically benchmarks all proposals, selects the winning candidate via machine markers (WINNER_CANDIDATE_INDEX: N), and promotes the winner's files directly into the project root directory.
4. Instant Live Canvas Previewing: When web applications or UI components are generated, dsh-moa integrates seamlessly with @goodandready/dsh-live-canvas, automatically spawning sandboxes for 1-click browser previewing.
5. Token-Saving Chat Summarization: Replaces massive code dumps in chat bubbles with compact file listings and clean architectural summaries.
6. One-Shot Session Model Restoration: Executes cleanly as a one-shot turn modifier, automatically reverting back to the user's primary session model immediately after completion.
7. Dynamic Model Pricing Catalog & Token Estimation: Real-time rate resolution for 300+ models fetched automatically in the background from OpenRouter's public catalog (cached locally in ~/.dsh/storages/dsh-moa-catalog.json for 24h), plus support for direct vendor rates and custom prices overrides in settings.yaml.
8. Refinement Mode (Incremental Edits): Automatically detects existing codebase context to generate precise delta modifications instead of destructive full-file rewrites.
9. Fast Mode & Custom Judge Criteria: Ultra-fast single-model preset for quick tasks and customizable evaluation guidelines for the judge.
10. Run History & Win-Rate Leaderboard: Persistent logging with built-in REST endpoints (/dsh-moa/history and /dsh-moa/leaderboard).

---

🏗️ Architecture

▼
mermaid
复制代码
graph TD
    subgraph Input ["User Interaction (Chat Composer)"]
        Cmd["Slash Command: /moa [preset] <prompt>"]
        Gate{"Ambiguity Check & Questionnaire"}
        QModal["Interactive Clarifying Questions<br/>(Options & Write-in responses)"]
    end

    subgraph Proposers ["Parallel Proposer Layer (Advisors)"]
        P1["Proposer Model 1<br/>(Creative Approach)"]
        P2["Proposer Model 2<br/>(Alternative Design)"]
        P3["Proposer Model 3<br/>(Performant Strategy)"]
        WS1[".moa/candidate-1/<br/>(Isolated Files)"]
        WS2[".moa/candidate-2/<br/>(Isolated Files)"]
        WS3[".moa/candidate-3/<br/>(Isolated Files)"]
    end

    subgraph Judge ["Synthesis & Promotion Layer"]
        Aggregator["Frontier Judge Model<br/>(Cross-Evaluation & Code Critique)"]
        WinnerMarker{"WINNER_CANDIDATE_INDEX"}
        Promote["Promote Winner Files<br/>(Move to project root & cleanup sandboxes)"]
        LiveCanvas["Live Canvas Integration<br/>(Auto-open Web UI sandbox)"]
        Summary["Token-Saving Summary<br/>(File overview & architecture highlights)"]
    end

    Cmd --> Gate
    Gate -->|Broad/Underspecified| QModal
    QModal -->|User Answers| P1 & P2 & P3
    Gate -->|Explicit/Detailed| P1 & P2 & P3
    P1 --> WS1
    P2 --> WS2
    P3 --> WS3
    WS1 & WS2 & WS3 --> Aggregator
    Aggregator --> WinnerMarker
    WinnerMarker --> Promote
    Promote --> LiveCanvas
    Promote --> Summary

---

✨ Features & Capabilities

1. Slash Command (/moa) & Autocompletion

Integrated directly into the DeepSeek Harness composer via client input triggers. Typing /moa shows presets and instant autocompletion:

▼
text
复制代码
/moa build a real-time reactive dashboard with charts and websocket updates

Or target a specific named preset:

▼
text
复制代码
/moa:code-review audit the auth middleware and security boundaries

2. Adaptive Questionnaire Gate

When prompts are open-ended or lack architectural specifications (e.g. "build a calculator app"), advisor models detect ambiguities and formulate focused clarifying questions (e.g., UI style, persistence backend, framework choice) before generating code.

3. Parallel Fan-Out with Live Heartbeats

• Proposers query concurrently with live heartbeat progress badges (⏳ [3s] Processing..., per-model completion status).
• Bulky system prompts and tool schemas are cleanly stripped from advisor contexts, eliminating "missing tools" refusals and token bloat.

4. Disk-Level Candidate Isolation & Promotion

Unlike standard chat-only MoA, dsh-moa isolates file generation onto the filesystem:

• Each proposer generates files into .moa/candidate-1/, .moa/candidate-2/, etc.
• The Judge compares implementations and selects the optimal solution with WINNER_CANDIDATE_INDEX: N.
• The winner's files are promoted to the workspace root, and temporary candidate directories are pruned automatically.

5. Live Canvas 1-Click Preview

If web files (index.html, React/JSX components, Vue, CSS) are generated, dsh-moa communicates with @goodandready/dsh-live-canvas via its REST endpoint to instantiate a live preview container with 1-click instant access.

6. Native Settings Card & Presets

Configure your models in Settings → Plugins → Mixture of Agents:

• Set custom Proposer models (e.g., fast generative models for diverse ideas).
• Set the Aggregator / Judge model (e.g., deep reasoning models for rigorous critique).
• Configure named presets (default, code-review, deep-reasoning).

---

📦 Installation

Install into your DeepSeek Harness web profile:

▼
bash
复制代码
dsh plugin --profile web add @goodandready/dsh-moa

Restart your DeepSeek Harness instance and refresh the browser.

---

⚙️ Configuration (settings.yaml)

Configure presets and model pipelines in settings.yaml or through the Web UI Settings panel:

▼
yaml
复制代码
# settings.yaml
dsh-moa:
  defaultPreset: "default"
  presets:
    default:
      references:
        - provider: "your-fast-provider"
          model: "your-creative-model"
        - provider: "your-fast-provider"
          model: "your-balanced-model"
      aggregator:
        provider: "your-reasoning-provider"
        model: "your-judge-model"
  prices:
    "my-provider/my-model":
      input: 0.20
      output: 0.80
    "ollama/*":
      input: 0
      output: 0
    code-review:
      references:
        - provider: "your-fast-provider"
          model: "your-security-model"
        - provider: "your-fast-provider"
          model: "your-performance-model"
      aggregator:
        provider: "your-reasoning-provider"
        model: "your-judge-model"

Configuration Parameters

Parameter	Type	Default	Description
defaultPreset	string	"default"	Default preset invoked when typing /moa <prompt>
presets.<name>.references	array	[...]	List of proposer models queried concurrently during the proposal phase
presets.<name>.aggregator	object	{...}	Frontier judge model responsible for synthesis, critique, and winner selection
enableQuestionnaire	boolean	true	Enable interactive clarifying questionnaire for underspecified requests
autoPromoteWinner	boolean	true	Automatically promote the judge's selected winner files into the project workspace

---

📊 REST API & Endpoints

Endpoint	Method	Description
/dsh-moa/presets	GET	Returns list of configured MoA presets
/dsh-moa/history?limit=20&offset=0	GET	Returns recent MoA runs with candidates, winner, cost, and tokens
/dsh-moa/leaderboard	GET	Computes model win-rate leaderboard and average execution costs

---

🧪 Testing

Run the automated test suite:

▼
bash
复制代码
npm test

---

📄 License

MIT © GooDAnDReaDY