Semantic Search MCP Server

{"id": "django_request_response_lifecycle", "description": "Full request/response lifecycle through middleware and views", "repo_path": "django", "query": "I'm debugging a weird issue where my response headers are getting modified somewhere. Can you trace through what happens to a request from when it hits Django to when the response goes out? I need to understand the full pipeline, not just middleware.", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "django/core/handlers/base.py", "must_include_substrings": ["def get_response(", "_middleware_chain", "def _get_response("]}, {"file_path": "django/core/handlers/wsgi.py", "must_include_substrings": ["class WSGIHandler", "def __call__(self, environ"]}]} {"id": "django_queryset_lazy_evaluation", "description": "How QuerySet defers database execution until iteration", "repo_path": "django", "query": "my colleague says querysets are lazy but I don't get it - when does django actually hit the database? I have a chain of filter().exclude().order_by() and I'm not sure when the SQL runs", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "django/db/models/query.py", "must_include_substrings": ["def __iter__(self)", "_fetch_all(", "_result_cache"]}, {"file_path": "django/db/models/sql/compiler.py", "must_include_substrings": ["def execute_sql("]}]} {"id": "gin_route_tree_matching", "description": "How Gin's radix tree matches URL paths to handlers", "repo_path": "gin", "query": "gin routing is fast but I don't understand how it finds my handler when I have hundreds of routes with parameters like /users/:id/posts/:postId - what's the data structure and matching algorithm?", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "tree.go", "must_include_substrings": ["type node struct", "func (n *node) getValue(", "indices", "wildChild"]}, {"file_path": "gin.go", "must_include_substrings": ["trees"]}]} {"id": "gin_context_pool_reuse", "description": "How Gin reuses Context objects via sync.Pool for performance", "repo_path": "gin", "query": "I heard gin reuses context objects for performance but that sounds dangerous - where does it manage context lifecycle and how does it make sure my handler gets a clean context?", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "gin.go", "must_include_substrings": ["pool", "sync.Pool", "func (engine *Engine) ServeHTTP("]}, {"file_path": "context.go", "must_include_substrings": ["func (c *Context) reset()"]}]} {"id": "codeql_dataflow_configuration", "description": "How CodeQL data flow configurations define sources, sinks and flow", "repo_path": "codeql", "query": "codeql found a vulnerability but the path it shows doesn't make sense to me - how does it actually build these source-to-sink paths? where's the configuration that defines what's a source and sink?", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "javascript/ql/lib/semmle/javascript/dataflow/Configuration.qll", "must_include_substrings": ["class Configuration", "isSource(", "isSink("]}, {"file_path": "javascript/ql/lib/semmle/javascript/dataflow/TaintTracking.qll", "must_include_substrings": ["TaintTracking"]}]} {"id": "codeql_security_query_structure", "description": "How security queries are structured to find vulnerabilities", "repo_path": "codeql", "query": "I want to write a custom codeql security query but don't understand the pattern - how do existing queries like XSS detection structure their source/sink/sanitizer logic?", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "javascript/ql/lib/semmle/javascript/security/dataflow/DomBasedXssQuery.qll", "must_include_substrings": ["DomBasedXss", "isSanitizer"]}, {"file_path": "javascript/ql/src/Security/CWE-079/DomBasedXss.ql", "must_include_substrings": ["from", "where", "select"]}]} {"id": "infinigen_procedural_material_nodes", "description": "How materials are procedurally generated using shader nodes", "repo_path": "infinigen", "query": "infinigen generates really complex materials but I can't figure out the pipeline - how does it go from random seeds to actual blender shader nodes? where's the node graph construction?", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "infinigen/core/nodes/node_wrangler.py", "must_include_substrings": ["class NodeWrangler", "new_node"]}, {"file_path": "infinigen/core/surface.py", "must_include_substrings": ["class NoApply", "apply"]}]} {"id": "infinigen_asset_placement_pipeline", "description": "How assets are placed and populated in scenes", "repo_path": "infinigen", "query": "I want to understand how infinigen puts together a full scene - like when it places trees, rocks, creatures together. Where does the high-level scene composition happen and how does it decide what goes where?", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "infinigen/core/placement/factory.py", "must_include_substrings": ["class AssetFactory"]}, {"file_path": "infinigen/core/execute_tasks.py", "must_include_substrings": ["execute_tasks"]}, {"file_path": "infinigen/core/constraints/example_solver/populate.py", "must_include_substrings": ["populate"]}]} {"id": "qgis_map_rendering_pipeline", "description": "How map layers are rendered in parallel to canvas", "repo_path": "QGIS", "query": "map rendering in qgis seems to involve multiple threads and caching - when I pan the map what's the sequence of calls from mouse event to pixels on screen? I need to understand the rendering architecture", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "src/core/maprenderer/qgsmaprendererparalleljob.cpp", "must_include_substrings": ["QgsMapRendererParallelJob", "start("]}, {"file_path": "src/gui/qgsmapcanvas.cpp", "must_include_substrings": ["refresh(", "QgsMapCanvas"]}]} {"id": "qgis_expression_evaluation", "description": "How QGIS evaluates expressions in symbology and labeling", "repo_path": "QGIS", "query": "I'm using data-defined expressions for styling but some are slow - where does qgis actually evaluate these expressions and how does the expression engine work internally?", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "src/core/expression/qgsexpression.cpp", "must_include_substrings": ["QgsExpression", "evaluate("]}, {"file_path": "src/core/expression/qgsexpressionnode.cpp", "must_include_substrings": ["eval("]}]} {"id": "claude_flow_agent_coordination", "description": "How multiple agents coordinate and communicate in swarm", "repo_path": "claude-flow", "query": "I have multiple agents running but they seem to step on each other's toes - where does claude-flow handle agent coordination and message passing between agents?", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "src/swarm/coordinator.ts", "must_include_substrings": ["SwarmCoordinator", "assignTask"]}, {"file_path": "src/swarm/claude-code-interface.ts", "must_include_substrings": ["ClaudeCodeInterface", "sendMessage"]}]} {"id": "claude_flow_memory_system", "description": "How agent memory persists and is retrieved across sessions", "repo_path": "claude-flow", "query": "agents in claude-flow seem to remember things across runs but I can't find where this memory is stored - is it in sqlite? files? how does the memory system work and where's the persistence layer?", "difficulty": "hard", "grep_resistant": true, "expected_items": [{"file_path": "src/swarm/memory.ts", "must_include_substrings": ["memory", "store"]}, {"file_path": "src/hive-mind/core/DatabaseManager.ts", "must_include_substrings": ["sqlite", "createTables"]}]}