Context+
Semantic Intelligence for Large-Scale Engineering.
Best for Performing deep code discovery in large, unfamiliar codebases.
Semantic Intelligence for Large-Scale Engineering.
Context+ is an MCP server designed for developers who demand 99% accuracy. By combining RAG, Tree-sitter AST, Spectral Clustering, and Obsidian-style linking, Context+ turns a massive codebase into a searchable, hierarchical feature graph.
What it does
- Hierarchical feature graph generation using Tree-sitter AST
- Semantic code search and navigation via spectral clustering
- Blast radius analysis for impact assessment of code changes
- Shadow restore points for safe AI-driven code modifications
- Graph-based memory management for codebase concepts and relations
Available tools (17)
get_context_treeStructural AST tree of a project with file headers and symbol ranges.get_file_skeletonFunction signatures, class methods, and type definitions with line ranges.semantic_code_searchSearch by meaning using embeddings over file headers and symbols.semantic_identifier_searchIdentifier-level semantic retrieval for functions, classes, and variables.semantic_navigateBrowse codebase by meaning using spectral clustering.get_blast_radiusTrace every file and line where a symbol is imported or used.run_static_analysisRun native linters and compilers to find unused variables, dead code, and type errors.propose_commitValidates code against strict rules before saving and creates a shadow restore point.get_feature_hubObsidian-style feature hub navigator for mapping features to code files.list_restore_pointsList all shadow restore points created by propose_commit.undo_changeRestore files to their state before a specific AI change.upsert_memory_nodeCreate or update a memory node with auto-generated embeddings.create_relationCreate typed edges between nodes.search_memory_graphSemantic search with graph traversal.prune_stale_linksRemove decayed edges and orphan nodes.add_interlinked_contextBulk-add nodes with auto-similarity linking.retrieve_with_traversalStart from a node and walk outward to return reachable neighbors.Setup requirements
Requires 1 environment variable: OLLAMA_EMBED_MODEL. Available via bunx and npx.