How does this skill prevent AI hallucinations during analysis?

It utilizes strict protocols including Assumption Validation Checkpoints and Evidence Chain Validation, requiring the AI to cite specific file paths and code snippets before making claims.

Can it help map out message bus interactions?

Yes, it features a dedicated Message Bus Analysis protocol that identifies producers and consumers across different services to map cross-service integration logic.

Does it create permanent documentation?

It generates structured Markdown analysis files in a temporary directory, which can be saved or used to generate final documentation for your team.

What frameworks is this investigation skill optimized for?

While it can investigate any codebase, it is specifically optimized for full-stack .NET and Angular architectures using patterns like CQRS, Clean Architecture, and Message Bus systems.

Feature Investigation

Name: Feature Investigation
Author: CongDon1207

byCongDon1207

•

Developer Tools

Analyzes and documents complex code logic flows, architectural patterns, and feature implementations across full-stack applications.

The Feature Investigation skill acts as a principal software architect to help developers rapidly master unfamiliar codebases or complex legacy features. It systematically traces execution paths from frontend UI components through API controllers and CQRS handlers down to domain entities and background jobs. By employing strict anti-hallucination protocols and structured external memory management, the skill builds a comprehensive knowledge graph of your project's logic, ensuring all explanations are grounded in verified code evidence rather than assumptions.

Key Features

01Deep tracing of message bus consumers, producers, and service dependencies

021 GitHub stars

03Automated construction of structured knowledge graphs for specific features

04Evidence-based anti-hallucination protocols for verified code analysis

05Standardized investigation reports including text-based architecture diagrams

06End-to-end logic flow mapping from frontend UI to backend persistence

Use Cases

01Onboarding to a new or legacy codebase by exploring existing feature implementations

02Documenting existing business logic for technical debt reduction and team knowledge sharing

03Debugging complex asynchronous workflows involving message brokers and background jobs

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add congdon1207/agents.md feature-investigation

For use in Claude.ai and ChatGPT

Download Skill