finwiz-context7

name	finwiz-context7
description	Automatically uses Context7 MCP tools to fetch up-to-date library documentation when working with external libraries. Use when implementing code with CrewAI, Pydantic, pytest, or any external library to ensure accuracy and compatibility.
allowed-tools	["mcp_context7_resolve_library_id","mcp_context7_get_library_docs","Read","Edit"]

FinWiz Context7 Integration

Key Principle: Proactively use Context7 MCP tools to fetch current library documentation. Don't wait for explicit requests - if you're implementing something with an external library, automatically get the docs.

When to Use Context7

Use Context7 tools automatically for:

• Code generation involving external libraries (CrewAI, Pydantic, pytest, etc.)
• Setup and configuration steps for dependencies
• Library/API documentation lookups for accurate implementation
• Version-specific features to ensure compatibility
• Best practices for library usage patterns

Context7 Workflow

Step 1: Resolve Library ID

// Use MCP tool to find the correct library
mcp_context7_resolve_library_id("crewai")
// Returns: "/joaomdmoura/crewai"

Step 2: Get Library Documentation

// Fetch relevant documentation
mcp_context7_get_library_docs(
  "/joaomdmoura/crewai", 
  topic="flow state management",
  tokens=5000
)

Step 3: Implement Using Current Patterns

Use the fetched documentation to implement code with:

• Current API methods (not deprecated ones)
• Proper parameter names and types
• Best practices from the library maintainers
• Version-compatible features

FinWiz-Specific Libraries

Common libraries that benefit from Context7 lookup:

Library	Context7 ID	Common Topics
CrewAI	/joaomdmoura/crewai	flow, agents, tasks, crews
Pydantic	/pydantic/pydantic	validation, strict mode, v2
pytest	/pytest-dev/pytest	fixtures, markers, parametrize
pytest-mock	/pytest-dev/pytest-mock	mocker fixture, patching
httpx	/encode/httpx	async client, authentication
pandas	/pandas-dev/pandas	dataframes, operations
FastAPI	/tiangolo/fastapi	endpoints, dependencies

Example Usage Patterns

CrewAI Flow Implementation

// 1. Resolve library
mcp_context7_resolve_library_id("crewai")

// 2. Get Flow documentation
mcp_context7_get_library_docs(
  "/joaomdmoura/crewai", 
  topic="flow state management pydantic models",
  tokens=7000
)

// 3. Implement using current patterns

Pydantic Model Validation

// Get current Pydantic v2 patterns
mcp_context7_get_library_docs(
  "/pydantic/pydantic",
  topic="strict mode validation extra forbid",
  tokens=4000
)

Testing with pytest-mock

// Get current pytest-mock patterns (unittest.mock is BANNED)
mcp_context7_get_library_docs(
  "/pytest-dev/pytest-mock",
  topic="mocker fixture patching best practices",
  tokens=3000
)

Benefits

Accuracy

• Current APIs: Use latest library methods, not deprecated ones
• Correct Parameters: Get proper parameter names and types
• Version Compatibility: Ensure code works with installed versions

Efficiency

• Avoid Errors: Prevent using outdated or incorrect patterns
• Best Practices: Follow library maintainer recommendations
• Faster Development: Get accurate info immediately

Compliance

• Library Standards: Follow official library patterns
• FinWiz Standards: Combine with existing FinWiz steering rules
• Quality Assurance: Reduce bugs from incorrect usage

Integration with FinWiz Standards

Context7 documentation should be used in conjunction with FinWiz steering rules:

• Validate patterns against finwiz-crewai skill
• Ensure testing follows finwiz-testing skill
• Apply validation per finwiz-validation skill
• Maintain quality per finwiz-development skill

Proactive Usage Examples

When You See These Patterns, Use Context7:

# Implementing CrewAI Flow → Get CrewAI docs
class MyFlow(Flow[StateModel]):
    # Auto-fetch CrewAI Flow documentation

# Creating Pydantic models → Get Pydantic docs  
class MyModel(BaseModel):
    # Auto-fetch Pydantic v2 documentation

# Writing tests with mocker → Get pytest-mock docs
def test_example(mocker):
    # Auto-fetch pytest-mock documentation

# Making HTTP requests → Get httpx docs
async with httpx.AsyncClient() as client:
    # Auto-fetch httpx documentation

Token Management

Adjust token limits based on complexity:

• Simple lookups: 3000 tokens
• Standard usage: 5000 tokens (default)
• Complex implementations: 7000-10000 tokens
• Comprehensive guides: 15000+ tokens

Error Handling

If Context7 lookup fails:

1. Continue with existing knowledge but note the limitation
2. Use FinWiz steering rules as fallback guidance
3. Document the assumption in code comments
4. Suggest manual verification if critical

Quality Assurance

After using Context7:

• Verify patterns match FinWiz standards
• Test implementation thoroughly
• Document any deviations from standard patterns
• Update FinWiz skills if new patterns emerge

Remember: Context7 provides the what (current library APIs), FinWiz skills provide the how (project-specific patterns and standards).