ruvnet/sublinear-time-solver

npx sublinear-time-solver mcp
npm install -g sublinear-time-solver
npm install sublinear-time-solver

Usage

• •npx sublinear-time-solver generate -t diagonally-dominant -s 1000 -o matrix.json
• •npx sublinear-time-solver solve -m matrix.json -b vector.json -o solution.json
• •npx sublinear-time-solver pagerank --graph graph.json --damping 0.85
• •Add {"command":"npx","args":["sublinear-time-solver","mcp"]} to Claude Desktop mcpServers

Key features

• ✓Sublinear solver suite: Neumann series, forward/backward push, hybrid random walk, and TRUE O(log n) Johnson-Lindenstrauss path
• ✓Auto method selection based on matrix properties (diagonal dominance, condition number)
• ✓WASM acceleration with complexity class declared at the API level
• ✓40+ unified MCP tools including PageRank, matrix analysis, and dynamic domain reasoning
• ✓Emergent self-modifying system, psycho-symbolic reasoning, and temporal/consciousness experiments
• ✓Nanosecond scheduler with hardware TSC timing

Best for

Developers working on large sparse matrix problems, PageRank, and graph analytics who want sublinear approximations in an LLM loop

Caveats

• ⚠Not ideal for small dense matrices, exact machine-precision solutions, or ill-conditioned systems
• ⚠O(log n) guarantees require diagonally dominant matrices
• ⚠Consciousness/temporal claims are experimental research material
• ⚠v1.6.0 fixed a CWE-73 arbitrary file write in the MCP tools — upgrade before use