Agent governance
Enterprise MCP Gateway
Register tenant-owned MCP upstreams, discover tools, enforce method and tool allowlists, bound execution, and retain metadata-only audit evidence.
[ Maha Strategies // Developer infrastructure ]
Build governed AI systems on an inspectable control layer.
One entry point for Maha's production APIs, SDKs, operational boundaries, benchmarks, and security model.
Use only the capability your workflow needs: govern agent tools, compile bounded context, evaluate evidence retention, run GPU optimization jobs, or preflight consequential documents.
[ Production capability map ]
Choose the control your workflow is missing.
Context engineering
Context Compiler
Rank and deduplicate task-relevant passages into a declared context budget while preserving source IDs, passage hashes, and measurable coverage.
Evaluation
Context Pack Evaluator
Declare evidence that must survive compilation, then measure exact retention alongside token, byte, coverage, and duplicate-removal metrics.
GPU optimization
Bounded optimization APIs
Run accurately bounded tensor-network QUBO/Ising heuristics and weighted SE(3) rigid registration through the asynchronous GPU job pipeline.
Evidence assurance
MPS Preflight
Screen AI-assisted documents before review and produce an explicit record of claims, evidence status, uncertainty, and remaining verification work.
[ SDKs and integration contracts ]
Integrate without adopting another runtime.
TypeScript SDK
Zero-dependency client for Node.js, Bun, Deno, browsers, and Edge runtimes.
Install @mahastrategies/sdk ↗Python SDK
Python client with optional LangChain and CrewAI adapters.
Install maha-sdk ↗OpenAPI 3.1
Machine-readable endpoint, schema, authentication, and error contracts.
Open the API contract ↗MCP bridge
Connect local agents to documented Maha APIs with explicit human approval for checkout.
Read the MCP bridge guide ↗[ Published operational evidence ]
Benchmarks state the boundary, not just the number.
The reported GPU measurements are seven-run warm solver baselines on an NVIDIA A10G. They do not include HTTP latency or container cold starts, and heuristic results do not claim global optimality.
Tensor-network QUBO / Ising
80.840 ms
Warm p95 at 256 variables and bond dimension 256; reviewed promotion threshold ≤150 ms.
Review scope and limits ↗Weighted SE(3) registration
108.604 ms
Warm p95 at 16,384 paired points; reviewed promotion threshold ≤200 ms.
Review scope and limits ↗Context compiler standard envelope
41.7 ms
Measured compute p95 at approximately 150,076 input tokens with BM25 and compound tokenization.
Review scope and limits ↗Evidence retention
Explicit
Declare required source spans and receive retained-or-omitted results instead of inferring quality from compression ratio.
Review scope and limits ↗[ Security and data boundary ]
Minimize what crosses and what remains.
Security claims are capability-specific. Review the API contract and deployment boundary before sending production, regulated, personal, or confidential data.
- Tenant isolation: API credentials resolve to tenant-scoped resources; one tenant cannot enumerate another tenant's MCP servers.
- Bounded execution: request sizes, methods, tools, rates, timeouts, failure thresholds, and circuit cooldowns are server-validated.
- Credential handling: supported MCP upstream secrets are encrypted at rest and are never returned by listing or discovery APIs.
- Data minimization: context compilation returns source text and the compiled pack transiently; the gateway audit record excludes tool arguments and upstream response bodies.