12 tools from the Oraclaw MCP Server, categorised by risk level.
View the Oraclaw policy →analyze_graph Graph analytics: PageRank, Louvain communities, shortest path, bottleneck detection. analyze_risk Portfolio risk: VaR/CVaR with correlation matrices. Monte Carlo simulation. detect_anomaly Anomaly/outlier detection (Z-score / IQR). Sub-millisecond. optimize_cmaes CMA-ES continuous optimization. Tune parameters, calibrate models. 10-100x more efficient than grid search. optimize_contextual Contextual Bandit (LinUCB). Context-aware selection — learns which option works best in which situation. predict_forecast Time series forecasting (ARIMA / Holt-Winters). Predict future values with confidence intervals. score_convergence Multi-source agreement scoring. How much do different signals/sources agree? simulate_montecarlo Monte Carlo simulation. 5K iterations in ~1ms. Risk quantification, scenario analysis. solve_constraints LP/MIP/QP optimization (HiGHS). Budget allocation, scheduling, resource planning. Provably optimal. optimize_bandit Multi-Armed Bandit (UCB1/Thompson/ε-Greedy). Select the best option from a set — optimal explore/exploit tradeoff. <1ms. 2/5 plan_pathfind A* pathfinding with k-shortest paths (Yen's algorithm). Optimal routing. 2/5 solve_schedule Optimal task scheduling with energy matching. Assigns tasks to time slots maximizing productivity. 2/5 The Oraclaw MCP server exposes 12 tools across 2 categories: Read, Write.
Use Intercept, the open-source MCP proxy. Write YAML rules for each tool — rate limits, argument validation, or deny rules — then run Intercept in front of the Oraclaw server.
Oraclaw tools are categorised as Read (9), Write (3). Each category has a recommended default policy.
Open source. One binary. Zero dependencies.
npx -y @policylayer/intercept