"author:Alpha-Park" matching MCP servers:

Enables statistical A/B testing by computing Welch's two-sample t-test, Satterthwaite degrees of freedom, and two-tailed p-values, alongside time-series forecasting, anomaly detection, regression, and PCA through MCP tools.

Enables agents to orchestrate and call tools across HubX's mobile AI app portfolio for multi-model chat, plant/vision diagnostics, generative image and design, nutrition tracking, audio note summarization, language tutoring, and wellness workflows through a unified MCP interface with structured JSON output.

Enables statistical forecasting and anomaly detection over numerical time-series data using Holt linear smoothing, multi-step projections, and variance-based confidence bands.

Enables MCP-compatible agents to audit reasoning trajectories for loops, deadlocks, and logical contradictions and to perform non-destructive checkpoint rollbacks.

Enables AI agents to fuse 2D/3D visual bounding boxes, user head-pose gaze vectors, and acoustic beamforming azimuth angles into cross-modal focal salience scores. It also resolves deictic spatial references in real time, running as a zero-dependency Python MCP server over JSON-RPC stdio.

Enables edge AI agents to optimize inference memory and latency by mapping logical blocks to non-contiguous physical pages, tracking fragmentation, and exposing related MCP tools.

Enables agents to perform symmetric INT8 per-tensor and per-channel quantization of LLM weights and activations, returning SNR and MSE reconstruction metrics for edge inference optimization.

Enables MCP-compliant clients and autonomous agents to pick optimal few-shot in-context exemplars by combining semantic token overlap, cosine similarity, and Maximal Marginal Relevance diversity ranking for prompt optimization. Runs deterministically on the Python standard library with zero external dependencies, avoiding frontier LLM round-trips for sub-millisecond selection.

Enables inference engines to match longest common prompt token prefixes via a radix trie, so cached KV-cache blocks are reused instead of re-prefilled. This eliminates redundant prefill computation and lowers time-to-first-token, alongside paged attention allocation, speculative decoding verification, INT8 quantization, and latency telemetry.

Enables MCP-compatible clients to deterministically neutralize adversarial prompt injections and jailbreak overrides with zero-dependency, sub-millisecond heuristic analysis.

Enables reversible anonymization of SSNs, credit cards, passport numbers, emails, and phone numbers in prompts before sending them to external LLMs, and restores original tokens from completions using local HMAC isolation.

Enables agents to verify complex math, code, and logic reasoning through MCTS step-level process reward modeling, counterfactual backtracking, symbolic proof checking, and majority-vote consensus so hallucinated branches get pruned. It also allocates test-time compute dynamically and exposes these capabilities over a zero-dependency STDIO MCP server for Claude Desktop, Cursor, and other agent runtimes.

Enables detecting LLM hallucinations and confabulations by clustering sampled responses into semantic equivalence classes and computing semantic entropy.

Enables users to perform multivariate linear regression with gradient descent, Holt linear forecasting, Welch's t-test evaluation, anomaly detection, and PCA dimensionality reduction through MCP-compatible clients using only Python standard library tools.

Enables agents to decay stale transient interactions and consolidate frequent episodic events into enduring semantic user persona traits using sleep-cycle and forgetting-curve inspired memory processing.

Enables autonomous agents to orchestrate HubX mobile AI tools for plant and nutrition vision diagnostics, voice transcription and summarization, photo-realistic portrait and art generation, and interior/exterior spatial design through structured JSON outputs.

Enables agents to deterministically shrink multi-turn conversation history and prompts by stripping verbose boilerplate and filler tokens while preserving critical tool parameters and named entities. This lets users cut inference costs by roughly 40-60% without losing essential context, running either as a native MCP server over JSON-RPC stdio or as an importable zero-dependency Python module.

Enables agents to check propositional logic assertions and validate formal symbolic deductions through MCP, supporting rigorous step-level reasoning verification.

Enables MCP-compatible clients to run speculative decoding draft-model verification, calculating acceptance probabilities and returning structured execution telemetry with zero external dependencies.

Enables agents to detect outliers in numerical streams using multiple robust statistical strategies and to perform related analytics such as forecasting, regression, hypothesis testing, and dimensionality reduction via MCP.