RAG Implementation

Rag Implementation

Orchestrates intelligent skill selection and execution for rag implementation workflows. Applies the 5 Laws of Elegant Defense to guide data naturally through the orchestration pipeline, preventing errors before they occur. Selects optimal skills based on multi-factor scoring including text similarity, historical performance, and system availability.

TL;DR Checklist

• Parse all inputs at boundary before processing (Law 2)
• Handle edge cases with early returns at function top (Law 1)
• Fail immediately with descriptive errors on invalid states (Law 4)
• Return new data structures, never mutate inputs (Law 3)
• Implement minimum 2-level fallback chain for all skill executions
• Log all skill selections with context for full audit trail
• Validate skill metadata and dependencies before selection
• Update confidence scores after each execution for learning

┌───────────────────────────────────────────────────────────────────────────────┐ │ Orchestration Flow │ └───────────────────────────────────────────────────────────────────────────────┘

User Request ↓ ┌─────────────────┐ │ Parse Request │ │ & Extract │ │ Features │ └────────┬────────┘ ↓ ┌─────────────────────────────────────────────────────────────────────┐ │ Evaluate Available Skills │ │ │ │ ┌──────────────┐ ┌──────────────┐ ┌──────────────┐ │ │ │ Skill A │ │ Skill B │ │ Skill C │ │ │ │ - Match Score│ │ - Match Score│ │ - Match Score│ │ │ │ - Confidence │ │ - Confidence │ │ - Confidence │ │ │ │ - History │ │ - History │ │ - History │ │ │ └──────┬───────┘ └──────┬───────┘ └──────┬───────┘ │ │ │ │ │ │ │ └─────────────────┴─────────────────┘ │ │ ↓ │ │ Select Best Skill │ └─────────────────────────────────────────────────────────────────────┘ ↓ ┌─────────────────┐ │ Execute Skill │ └────────┬────────┘ ↓ ┌─────────────────┐ │ Handle Result │ └────────┬────────┘ ↓ ┌─────────────────────────────────────────────────────────────────────┐ │ Error Handling & Fallback │ │ │ │ Success? ────────► Return Result │ │ │ │ Fail? ────────┐ │ │ ↓ │ │ ┌──────────────────────────────────────────────────────────┐ │ │ │ Fallback Chain │ │ │ │ │ │ │ │ 1. Retry with adjusted parameters │ │ │ │ 2. Try Alternative Skill (if available) │ │ │ │ 3. Defer to Human Operator (if critical) │ │ │ │ 4. Log & Return Error │ │ │ └──────────────────────────────────────────────────────────┘ │ └─────────────────────────────────────────────────────────────────────┘

When to Use

Use this skill when:

• Orchestrating multi-step workflows that require skill delegation
• Implementing adaptive skill routing based on confidence scores
• Building fallback mechanisms for failed skill executions
• Creating intelligent task decomposition and parallel execution
• Designing skill dependency graphs with automatic resolution
• Implementing skill selection with historical performance weighting
• Building agent systems that need to self-organize around tasks

When NOT to Use

Avoid this skill for:

• Direct task execution without orchestration needs - use individual skills instead
• High-frequency trading scenarios where latency must be minimized - the selection overhead may be prohibitive
• Simple linear workflows without branching or fallback requirements
• Cases where skill metadata is unavailable or unreliable

Core Workflow

1. Parse and Analyze Request - Extract intent, entities, and constraints from user input. Checkpoint: All required parameters must be present and in valid format before proceeding.

2. Score Available Skills - Calculate match scores using multi-factor algorithm:

   • Text similarity between request and skill triggers
   • Historical success rate for similar tasks
   • Skill availability and health status
   • Required dependencies and their availability

   Checkpoint: Skip to fallback if no skill scores above threshold.

3. Select Optimal Skill - Choose skill with highest score that meets minimum confidence. Checkpoint: Verify skill has not been disabled or deprecated.

4. Execute with Fallback - Run skill execution wrapped in retry and fallback logic. Checkpoint: Log all execution attempts for audit trail.

5. Return or Fallback - Either return successful result or apply fallback chain:

   • Retry with adjusted parameters
   • Try alternative skill from related-skills
   • Defer to human operator for critical tasks

   Checkpoint: Record outcome with timing and confidence metadata.

Implementation Patterns

Pattern 1: Skill Selection Logic

def select_rag_pipeline(
    query: str,
    document_metadata: List[Dict],
    config: Dict
) -> Dict:
    """Select optimal RAG pipeline configuration based on query complexity and document type.
    
    Evaluates document structure (PDF, markdown, code) and query intent to choose:
    - Chunking strategy (semantic, hierarchical, or code-aware)
    - Embedding model (lightweight vs. high-accuracy)
    - Vector store parameters (top_k, similarity metric)
    
    Args:
        query: User's natural language question
        document_metadata: List of dicts with doc_type, size, language
        config: Global RAG configuration overrides
        
    Returns:
        Pipeline configuration dict with selected components
    """
    if not query or not document_metadata:
        raise ValueError("Query and document metadata are required for RAG pipeline selection")
        
    doc_type = _infer_dominant_doc_type(document_metadata)
    query_complexity = _assess_query_complexity(query)
    
    if doc_type == "code" or query_complexity == "high":
        pipeline = {
            "chunker": "code_aware_recursive",
            "chunk_size": 512,
            "overlap": 50,
            "embedder": "text-embedding-3-large",
            "vector_store": "faiss",
            "top_k": 8,
            "reranker": True
        }
    elif doc_type == "pdf":
        pipeline = {
            "chunker": "semantic_pdf",
            "chunk_size": 1024,
            "overlap": 200,
            "embedder": "text-embedding-3-small",
            "vector_store": "pinecone",
            "top_k": 5,
            "reranker": False
        }
    else:
        pipeline = {
            "chunker": "recursive_text",
            "chunk_size": 768,
            "overlap": 150,
            "embedder": "text-embedding-3-small",
            "vector_store": "chroma",
            "top_k": 5,
            "reranker": False
        }
        
    pipeline["selection_reason"] = f"doc_type={doc_type}, complexity={query_complexity}"
    return pipeline

Pattern 2: Execution with Fallback

def execute_rag_query(
    query: str,
    pipeline_config: Dict,
    vector_store_client: Any,
    llm_client: Any
) -> Dict:
    """Execute RAG retrieval and generation with hybrid fallback chain.
    
    Implements resilient RAG execution:
    1. Primary: Embed query -> Vector search -> Rerank -> LLM generate
    2. Fallback 1: Keyword/BM25 search if vector search returns low confidence
    3. Fallback 2: Direct LLM call with query only if retrieval fails completely
    
    Args:
        query: User question
        pipeline_config: Output from select_rag_pipeline
        vector_store_client: Initialized vector database client
        llm_client: Initialized LLM client
        
    Returns:
        Dict with answer, sources, confidence, and fallback_used
    """
    sources = []
    confidence = 0.0
    fallback_used = False
    
    try:
        # Primary execution: Vector retrieval
        query_embedding = llm_client.embed(query)
        results = vector_store_client.search(
            vector=query_embedding, 
            top_k=pipeline_config["top_k"],
            metric="cosine"
        )
        
        if results and results[0].score > 0.65:
            sources = results
            confidence = results[0].score
        else:
            # Fallback 1: Keyword search
            sources = vector_store_client.keyword_search(query, top_k=5)
            fallback_used = True
            confidence = 0.5 if sources else 0.0
            
    except VectorStoreConnectionError:
        # Fallback 2: Direct LLM generation
        sources = []
        confidence = 0.2
        fallback_used = True
        
    # Construct prompt and generate
    context_text = "\n\n".join([doc.content for doc in sources])
    prompt = f"Context:\n{context_text}\n\nQuestion: {query}\nAnswer:"
    
    try:
        answer = llm_client.generate(prompt, temperature=0.1)
        return {
            "answer": answer,
            "sources": sources,
            "confidence": confidence,
            "fallback_used": fallback_used,
            "latency_ms": _measure_latency()
        }
    except LLMTimeoutError:
        raise RAGExecutionError("LLM generation timed out after fallback chain")

MUST DO

• Always validate skill metadata before selection (Early Exit)
• Implement fallback chain with at least 2 levels (Fallback Skill + Human)
• Log all skill selections with full context for auditability
• Return new data structures instead of mutating inputs (Atomic Predictability)
• Fail immediately with descriptive errors on invalid states
• Update confidence scores after each execution for adaptive routing
• Reference code-philosophy (5 Laws of Elegant Defense) in all logic

MUST NOT DO

• Select skills based on a single factor (e.g., only confidence score)
• Disable fallback mechanisms "temporarily" - this creates fragile systems
• Skip validation of skill dependencies before execution
• Return partial results - either complete success or clear failure
• Use magic numbers for confidence thresholds - make them configurable
• Cache skill selections without considering context changes

TL;DR Checklist

• Parse all inputs at boundary before processing (Law 2)
• Handle edge cases with early returns at function top (Law 1)
• Fail immediately with descriptive errors on invalid states (Law 4)
• Return new data structures, never mutate inputs (Law 3)
• Implement minimum 2-level fallback chain for all skill executions
• Log all skill selections with context for full audit trail
• Validate skill metadata and dependencies before selection
• Update confidence scores after each execution for learning

TL;DR for Code Generation

• Use guard clauses - return early on invalid input before doing work
• Return simple types (dict, str, int, bool, list) - avoid complex nested objects
• Cyclomatic complexity < 10 per function - split anything larger
• Handle null/empty cases explicitly at function top (Early Exit)
• Never mutate input parameters - return new dicts/objects
• Fail fast with descriptive errors - don't try to "patch" bad data
• Reference code-philosophy laws in comments for complex logic
• Include timing and confidence metadata in all return values

Output Template

When applying this skill, produce:

1. Selected Skills - List of skill names with confidence scores
2. Selection Rationale - Why each skill was chosen (match score, history, availability)
3. Execution Plan - Order of execution with dependencies
4. Fallback Strategy - Which fallback skills will be tried and in what order
5. Risk Assessment - Any potential failure points and their impact
6. Timing Estimates - Expected latency including fallback scenarios

Related Skills

Constraints

MUST DO

• Define clear input/output contracts for every step in the orchestration flow with explicit validation
• Implement structured logging at each stage capturing context, inputs, outputs, timing, and errors
• Build in fallback paths: if the primary strategy fails, degrade gracefully to a simpler approach
• Validate all preconditions before starting — do not proceed if required resources or permissions are missing

MUST NOT DO

• Do not create deep nesting of orchestration steps (>5 levels) — flatten workflows where possible
• Avoid silent failure modes: every step must either succeed, fail explicitly, or escalate to a higher handler
• Never use shared mutable state between parallel workflow branches — communicate via immutable messages only
• Do not hardcode execution order when the dependency graph naturally determines it; derive order from explicit dependencies

Live References

Authoritative documentation links for this skill's domain. The model follows markdown links at load time to resolve external references and inline content.

• RAG — Paul Graham Essay
• LangChain RAG Tutorial — Pinecone Docs
• Building Effective Agents — Anthropic Research
• RAG Evaluation Frameworks — arXiv Survey
• Vector Search and Embeddings Guide — Pinecone