$ loading_
Local-first MCP server for reconstructing architecture, workflows, and business rules from code.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "io.github.silex-tec/silex" MCP server from askskill: Run: claude mcp add 'io-github-silex-tec-silex' -- npx -y silex-archaeology
Analyze this codebase, reconstruct the system architecture, and explain the relationships between its modules.
An architecture overview with key modules, dependencies, and a description of the overall structure.
Identify the core business rules from the code, group them by functional domain, and note the related implementation locations.
A categorized list of business rules with code clues or module locations.
Infer the main workflows from the code and describe the key steps, inputs and outputs, and involved components.
A workflow description showing how the system completes its core business operations.
Developers can use it to reconstruct architecture and workflows directly from code when documentation is missing. It also helps surface key business rules during handoff.
Product managers can use it to understand workflows and business rules from the implementation side. It is useful when they need a faster view of how an existing system behaves.
Researchers or technical leads can use it to extract structured understanding from large codebases for review, analysis, or knowledge capture. Its focus is reconstructing system knowledge from code.
It is a local-first MCP server that reconstructs system architecture, workflows, and business rules from code. It is designed to help users understand existing codebases.
Based on the description, it works from your code to perform reconstruction and analysis. For exact integration steps or input formats, see the source repository.
The description explicitly says it is “local-first.” For whether it is fully offline and for runtime requirements, see the source repository.
Manage project architecture, modules, and data flows locally with full privacy.
Get technical advice, code reviews, and explanations via Codex CLI.
Give AI coding assistants memory, code graph insight, and safe multi-agent coordination.
Query a local repo index for deterministic code search and analysis.
Analyze codebases with semantic search, dependency insights, and natural language Q&A.
Delegate coding tasks to Codex via MCP with security and result checks.