dspy-skills MCP — Casos de uso, Instalar & Demo ao vivo

Por que usar

Principais recursos

DSPy signatures / modules / predictors encoded as patterns
Teleprompter (BootstrapFewShot, MIPRO) usage guidance
Evaluation + metric design
RAG pipeline scaffolds with DSPy retrievers
Opinionated integration with common LLM providers

Demo ao vivo

Como fica na prática

dspy-skill.replay ▶ pronto

0/0

Instalar

Escolha seu cliente

~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "dspy-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/OmidZamani/dspy-skills",
        "~/.claude/skills/dspy-skills"
      ],
      "_inferred": true
    }
  }
}

Abra Claude Desktop → Settings → Developer → Edit Config. Reinicie após salvar.

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "dspy-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/OmidZamani/dspy-skills",
        "~/.claude/skills/dspy-skills"
      ],
      "_inferred": true
    }
  }
}

Cursor usa o mesmo esquema mcpServers que o Claude Desktop. Config de projeto vence a global.

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "dspy-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/OmidZamani/dspy-skills",
        "~/.claude/skills/dspy-skills"
      ],
      "_inferred": true
    }
  }
}

Clique no ícone MCP Servers na barra lateral do Cline, depois "Edit Configuration".

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "dspy-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/OmidZamani/dspy-skills",
        "~/.claude/skills/dspy-skills"
      ],
      "_inferred": true
    }
  }
}

Mesmo formato do Claude Desktop. Reinicie o Windsurf para aplicar.

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "dspy-skill",
      "command": "git",
      "args": [
        "clone",
        "https://github.com/OmidZamani/dspy-skills",
        "~/.claude/skills/dspy-skills"
      ]
    }
  ]
}

O Continue usa um array de objetos de servidor em vez de um map.

~/.config/zed/settings.json

{
  "context_servers": {
    "dspy-skill": {
      "command": {
        "path": "git",
        "args": [
          "clone",
          "https://github.com/OmidZamani/dspy-skills",
          "~/.claude/skills/dspy-skills"
        ]
      }
    }
  }
}

Adicione em context_servers. Zed recarrega automaticamente ao salvar.

claude mcp add dspy-skill -- git clone https://github.com/OmidZamani/dspy-skills ~/.claude/skills/dspy-skills

Uma linha só. Verifique com claude mcp list. Remova com claude mcp remove.

Casos de uso

Usos do mundo real: dspy-skills

How to build your first DSPy program and optimize it

👤 ML engineers and applied researchers ⏱ ~90 min advanced

Quando usar: You have a task where prompt quality matters and want a systematic way to improve it.

Pré-requisitos

Python 3.10+ with dspy-ai installed — pip install dspy-ai
Skill cloned — git clone https://github.com/OmidZamani/dspy-skills ~/.claude/skills/dspy-skills

Fluxo

Define the signature

Task: classify support tickets into {billing, technical, account}. Give me a DSPy signature and a simple Predict module.✓ Copiado

→ Signature + module code
Write an eval

Add an evaluation set of 50 labeled examples and an accuracy metric.✓ Copiado

→ Eval harness with metric callable
Optimize

Run BootstrapFewShot to compile the module against the eval set.✓ Copiado

→ Compiled predictor + improved score

Resultado: A DSPy-optimized predictor that beats a hand-written prompt, with reproducible code.

Armadilhas

Eval too small — optimizer overfits — Minimum 100–200 examples; hold out a true test set
Metric doesn't capture what you care about — Spend on metric design before on model choice

Combine com: filesystem

Build a RAG pipeline with DSPy

👤 Engineers building retrieval-augmented systems ⏱ ~120 min advanced

Quando usar: You want modular, optimizable RAG rather than a hand-wired chain.

Fluxo

Define the modules

Create a DSPy RAG pipeline: RetrieveThenRead with ColBERTv2 or a local retriever.✓ Copiado

→ Modular pipeline with separate retrieval and generation
Optimize end-to-end

Write an eval on our QA set and run MIPRO to improve.✓ Copiado

→ Compiled pipeline with score delta

Resultado: A RAG pipeline you can improve by changing the eval, not by rewriting prompts.

Armadilhas

Retriever quality caps end-to-end quality — Evaluate retrieval separately (recall@k) before optimizing generation

Combine com: local-rag

Combinações

Combine com outros MCPs para 10× de alavancagem

dspy-skill + local-rag

Plug a local retriever into DSPy's RAG modules

Swap the ColBERT retriever for a local-rag MCP as the retrieval source.✓ Copiado

dspy-skill + filesystem

Organize DSPy programs, evals, and artifacts in a repo

Lay out a DSPy project with programs/, evals/, and artifacts/ directories.✓ Copiado

Ferramentas

O que este MCP expõe

Ferramenta	Entradas	Quando chamar	Custo
signature-design	task spec	Start of any DSPy program	0
module-authoring	signatures + flow	After signatures	0
teleprompter-optimization	module + eval	After eval is ready	LLM tokens during optimization
evaluation-harness	task data	Before optimizing	0

Custo e limites

O que custa rodar

Cota de API: LLM tokens dominate during optimization runs
Tokens por chamada: Can be high — optimizations may invoke the LLM hundreds of times
Monetário: depends on provider
Dica: Use cheap models during teleprompter runs, upgrade only for final evaluation

Segurança

Permissões, segredos, alcance

Armazenamento de credenciais: LLM provider keys in env vars

Saída de dados: LLM provider endpoints

Solução de problemas

Erros comuns e correções

Teleprompter seems to make things worse

Check metric correctness; use a held-out set; widen the example pool.

Optimization eats your budget

Cap max_bootstrapped_demos and use cheaper models during search.

Alternativas

dspy-skills vs. outros

Alternativa	Quando usar	Troca
prompt-architect-skill	You want prompt-level craft, not DSPy's programmatic approach	Hand-crafted vs optimized

Mais

Recursos

📖 Leia o README oficial no GitHub

🐙 Ver issues abertas

🔍 Ver todos os 400+ servidores MCP e Skills