OraClaw Decision Intelligence

12 MCP tools with 19 ML algorithms for AI agents — bandits, solvers, forecasters, risk models. All under 25ms, deterministic.

GitHub

OraClaw

MIT License Tests MCP Algorithms Latency npm API Status Implementations

MCP Optimization Tools for AI Agents -- 17 tools, 21 algorithms, sub-25ms. Zero LLM cost.

Your AI agent can't do math. OraClaw gives it deterministic optimization, simulation, forecasting, and risk analysis through the Model Context Protocol. Every tool returns structured JSON, runs in under 25ms, and costs nothing to compute.

Validation

OraClaw's math has been independently implemented in 12 open-source projects across AI agent orchestration, time-series tracking, vector search, MIP optimization, and production ML systems -- all within the first 8 days after public launch.

Selected field implementations (see CHANGELOG.md for the full list):

  • chernistry/bernstein -- 84⭐ agent orchestration framework. LinUCB contextual router with α=0.3, shadow-evaluation path, interpretable decision reasons. Shipped in codex/issue-367-linucb-router 1h40m after the spec correction.
  • stxkxs/nanohype -- contextual bandit routing, pluggable strategy registry (hash / sliding-TTL / semantic), cost anomaly detection, LinUCB on roadmap. "Your input shaped a lot of what actually shipped."
  • rfivesix/hypertrack -- Bayesian/Kalman-style adaptive calorie estimator with phase-aware kcal/kg ramp. Shipped in 0.8.0-beta. "At this point I think the mathematical model is in a very strong place."
  • AlanHuang99/pyrollmatch -- entropy balancing (Hainmueller 2012) with moment constraints + max_weight cap. Shipped in v0.1.3.
  • stffns/vstash -- IDF-sigmoid relevance weighting. Shipped in v0.17.0.

Marketplace distribution:

Maintainer relationships (warm technical correspondence): Qdrant, Milvus, NetworkX, Apache DataFusion, DuckDB, pymc-labs.

Quick Start

1. MCP Server (recommended for AI agents)

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "oraclaw": {
      "command": "npx",
      "args": ["-y", "@oraclaw/mcp-server"]
    }
  }
}

Then ask your agent:

"I have 3 email subject line variants. Which should I send next?"

The agent calls optimize_bandit and gets a statistically optimal selection in 0.01ms.

2. REST API (no install)

curl -X POST https://oraclaw-api.onrender.com/api/v1/optimize/bandit \
  -H 'Content-Type: application/json' \
  -d '{
    "arms": [
      {"id": "A", "name": "Option A", "pulls": 10, "totalReward": 7},
      {"id": "B", "name": "Option B", "pulls": 10, "totalReward": 5},
      {"id": "C", "name": "Option C", "pulls": 2, "totalReward": 1.8}
    ],
    "algorithm": "ucb1"
  }'

Response (<1ms):

{
  "selected": { "id": "C", "name": "Option C" },
  "score": 1.876,
  "algorithm": "ucb1",
  "exploitation": 0.9,
  "exploration": 0.976,
  "regret": 0.1
}

Free tier: 25 calls/day, no API key needed.

3. npm SDK

npm install @oraclaw/bandit

import { OraBandit } from '@oraclaw/bandit';

const client = new OraBandit({ baseUrl: 'https://oraclaw-api.onrender.com' });
const result = await client.optimize({
  arms: [
    { id: 'A', name: 'Short Subject', pulls: 500, totalReward: 175 },
    { id: 'B', name: 'Long Subject', pulls: 300, totalReward: 126 },
  ],
  algorithm: 'ucb1',
});

14 SDK packages: @oraclaw/bandit, @oraclaw/solver, @oraclaw/simulate, @oraclaw/risk, @oraclaw/forecast, @oraclaw/anomaly, @oraclaw/graph, @oraclaw/bayesian, @oraclaw/ensemble, @oraclaw/calibrate, @oraclaw/evolve, @oraclaw/pathfind, @oraclaw/cmaes, @oraclaw/decide

Why?

LLMs generate plausible text, not optimal solutions. Ask GPT to pick the best A/B test variant and it applies a heuristic that ignores the exploration-exploitation tradeoff. Ask it to solve a linear program and it hallucinates constraints. OraClaw gives your agent access to real algorithms -- bandits, solvers, forecasters, risk models -- that return mathematically correct answers in sub-millisecond time, without burning tokens on reasoning.

MCP Tool Catalog (17 tools)

Free tier (11 tools, no API key — 25 calls/day per IP):

ToolWhat It DoesLatency
optimize_banditUCB1 / Thompson / Epsilon-Greedy arm selection0.01ms
optimize_contextualContext-aware LinUCB bandit0.05ms
optimize_evolveGenetic algorithm for discrete + multi-objective problems<10ms
solve_scheduleEnergy-matched task scheduling3ms
score_convergenceMulti-source probability consensus (Hellinger)0.04ms
score_calibrationBrier + log score for forecaster accuracy0.02ms
predict_bayesianBeta posterior update from weighted evidence0.05ms
predict_ensembleMulti-model consensus + uncertainty decomposition0.1ms
plan_pathfindA* + Yen's k-shortest paths0.1ms
simulate_montecarloSingle-factor Monte Carlo (6 distributions)<2ms
simulate_scenarioWhat-if comparison + sensitivity ranking<5ms

Premium tier (6 tools, requires ORACLAW_API_KEY):

ToolWhat It DoesLatency
optimize_cmaesCMA-ES continuous black-box optimization12ms
solve_constraintsLP / MIP / QP solver via HiGHS (provably optimal)2ms
analyze_graphPageRank, Louvain communities, bottleneck detection0.5ms
analyze_riskVaR and CVaR (Expected Shortfall)<2ms
predict_forecastARIMA + Holt-Winters time series forecasting0.08ms
detect_anomalyZ-Score + IQR anomaly detection0.01ms

14 of 18 REST endpoints respond in under 1ms. All under 25ms.

Try It Now

The API is live. No signup required.

# Bayesian inference
curl -X POST https://oraclaw-api.onrender.com/api/v1/predict/bayesian \
  -H 'Content-Type: application/json' \
  -d '{"prior": 0.3, "evidence": [{"factor": "positive_test", "weight": 0.9, "value": 0.05}]}'

# Monte Carlo simulation
curl -X POST https://oraclaw-api.onrender.com/api/v1/simulate/montecarlo \
  -H 'Content-Type: application/json' \
  -d '{"simulations": 1000, "distribution": "normal", "params": {"mean": 100, "stddev": 15}}'

# Anomaly detection
curl -X POST https://oraclaw-api.onrender.com/api/v1/detect/anomaly \
  -H 'Content-Type: application/json' \
  -d '{"data": [10, 12, 11, 13, 50, 12, 11, 10], "method": "zscore", "threshold": 2.0}'

Pricing

TierCallsPriceAuth
Free25/day$0None
Pay-per-call1K/day$0.005/callAPI key
Starter10K/mo$9/moAPI key
Growth100K/mo$49/moAPI key
Scale1M/mo$199/moAPI key

x402 USDC: AI agents pay $0.01-$0.15 per call with USDC on Base. No subscription, no API key.

Source Code

ComponentPath
MCP Servermission-control/packages/mcp-server/
REST APImission-control/apps/api/
Algorithmsmission-control/apps/api/src/services/oracle/algorithms/
SDK Packagesmission-control/packages/sdk/
LangChain Toolsmission-control/integrations/langchain/oraclaw_tools.py
Mobile Appmission-control/apps/mobile/
Dashboard (Next.js)web/

Building with OraClaw?

We'd love to hear what you're working on. Share your use case, ask questions, or request features:

Links

If this saved your agent from hallucinating math, star us :star:

License

MIT

相關伺服器