LLM Mart Basic

Scalable deep-learning training with PyTorch Lightning — organize PyTorch code into LightningModules, configure Trainers for multi-GPU/TPU, build data pipelines and callbacks, log to W&B or TensorBoard, and run distributed training (DDP, FSDP, DeepSpeed). Use when structuring PyT

Classical machine learning in Python with scikit-learn — algorithms, preprocessing, pipelines, and best-practice reference documentation. Use when working with supervised learning (classification, regression), unsupervised learning (clustering, dimensionality reduction), model ev

Survival analysis and time-to-event modeling in Python with scikit-survival. Use when working with censored survival data, fitting Cox models, Random Survival Forests, Gradient Boosting models, or Survival SVMs, evaluating predictions with concordance index or Brier score, handli

Model interpretability and explainability with SHAP (SHapley Additive exPlanations) — feature importance and plots (waterfall, beeswarm, bar, scatter, force, heatmap). Use when explaining ML model predictions, computing feature importance, debugging models, analyzing bias or fair

Process-based discrete-event simulation in Python with SimPy — processes, queues, shared resources, and time-based events. Use when simulating systems where entities contend for shared resources over time, such as manufacturing systems, service operations, network traffic, or log

Trains single-agent reinforcement learning agents with Stable-Baselines3 — PPO, SAC, DQN, TD3, DDPG, and A2C behind a scikit-learn-like API. Use for standard single-agent RL experiments, quick prototyping, well-documented algorithm implementations on Gymnasium environments, or ad

Guided statistical analysis with hypothesis-test selection, assumption checking, effect sizes, power analysis, and APA-formatted reporting using scipy.stats, statsmodels, and pingouin (Bayesian alternatives with PyMC). Use when choosing and running the appropriate statistical tes

Statistical modeling in Python with statsmodels — OLS/WLS/GLS, GLM, discrete-choice and count models, mixed models, ARIMA/SARIMAX/VAR, with diagnostics, robust standard errors, and coefficient-level inference. Use when fitting specific model classes for econometrics, time series,

Forecasts time series zero-shot with Google's TimesFM foundation models — TimesFM 2.5 (200M, Apache-2.0 weights; ForecastConfig API, XReg covariates) and TimesFM 3.0 (~330M, multivariate with native past/future covariates; non-commercial weights) — producing point forecasts and q

Graph Neural Networks with PyTorch Geometric (PyG) — node and graph classification, link prediction, GCN, GAT, and GraphSAGE layers, heterogeneous graphs, and molecular property prediction. Use when building or training GNNs for geometric deep learning on graph-structured data. P

Loads, runs, and fine-tunes pretrained models with Hugging Face Transformers v5 (PyTorch-only) — pipeline() inference for chat-model text generation, text classification, NER, zero-shot, speech recognition, image classification, object detection, and image-text-to-text VLMs; Auto

Nonlinear dimensionality reduction with UMAP — fast manifold learning for 2D/3D visualization, clustering preprocessing (e.g., HDBSCAN), and supervised or parametric UMAP. Use when projecting high-dimensional data to low dimensions for visualization, embedding generation, or as a

Out-of-core tabular analytics with Vaex — memory-mapped HDF5/Arrow/Parquet via vaex.open, lazy virtual columns, delayed single-pass aggregations on billion-row tables, binned histograms/heatmaps, and vaex.ml transformers on one machine. Vaex is in minimal-maintenance mode (vaex-c

Access the AlphaFold DB of 240M+ AI-PREDICTED protein structures (v6, plus precomputed homodimer/heterodimer complexes) — retrieve models by UniProt accession, download PDB/mmCIF files, and analyze prediction confidence metrics (pLDDT, PAE). Use when a UniProt ID needs a computat

Search and retrieve preprints from arXiv via the Atom API by keywords, authors, arXiv IDs, date ranges, or subject categories. Use when finding or fetching papers in physics, mathematics, computer science, quantitative biology, quantitative finance, statistics, electrical enginee

Query BindingDB for measured protein-ligand binding affinities (Ki, Kd, IC50, EC50) via its keyless REST API or the full TSV download, searching by target (UniProt ID), compound (SMILES), or pathogen. Use when looking up experimental binding constants, profiling inhibitors of a p

Search the bioRxiv preprint server and retrieve paper metadata or download PDFs via its API. Use when finding life sciences preprints by keywords, authors, DOI, date ranges, or categories, or when conducting a biology literature review of not-yet-peer-reviewed work. Part of the A

Access the BRENDA enzyme database via its SOAP API to retrieve kinetic parameters (Km, kcat, Ki), reaction equations, organism data, and substrate-specific enzyme information indexed by EC number. Use when looking up enzyme kinetics, turnover numbers, or substrate specificity for

Query cBioPortal via its keyless REST API for cancer genomics across TCGA, GENIE, MSK-IMPACT and hundreds of studies — somatic mutations, copy-number alterations (GISTIC), mRNA/protein expression, structural variants, and patient-level clinical/survival data. Use when asked how o

Query ChEMBL via the chembl_webresource_client Python client for curated bioactive molecules and drug-like compound libraries at scale — search compounds by structure or physicochemical properties, retrieve bioactivity measurements (IC50, Ki, EC50), and find inhibitors of a targe

Fourteen posts of being wrong in production, compressed to checkboxes

Healthy nodes, a quiet network, 300 restarts in three days, and a latency budget measured in milliseconds

Discovery worked. Ping worked. Every TCP connection timed out, and later the tunnel only worked when someone had a terminal open.

Every VM came back. The cluster did not. Declarative systems converge on config, and the datapath isn't config.

A surprising share of AI-in-the-terminal failures aren't the AI. They're zsh, and a version of bash from 2006.

A Claude Code plugin turns standalone project configuration into a namespaced, installable extension that teams and communities can update as one unit.

None of the safety came from the model. It came from six boring habits.

Skills package instructions and references. Subagents run work in a separate context and return results. They solve different problems and can be composed deliberately.

Six hours in, one step left, everything green, and the incident that didn't happen

CLAUDE.md carries persistent project context. Skills load reusable procedures when relevant. Separating stable facts from task-specific workflows keeps both easier to maintain.

Twenty minutes recovering secrets that never existed, and the one sentence from a human that ended it

An API request routing a model's tool call through an approval gate to a remote MCP server

31 config keys, two audits, and why the first one was wrong in both directions

The official MCP Registry stores standardized server metadata rather than package code. Publishers verify a namespace, describe installation or remote access, and submit immutable versions.

Everyone looks at the Dockerfile. The file that actually leaked the key was the project file.

Remote MCP authorization uses established OAuth standards, but secure integration still requires issuer validation, least-privilege scopes, protected token handling, and server-side enforcement.

"Copy it over and switch the reference" is two steps, and the outage lives in the one nobody checks

stdio fits local processes and prototypes. Streamable HTTP fits hosted services and shared integrations. The right choice follows where the capability runs and who must reach it.

The most important rule wasn't about what I could change. It was about what I was allowed to display.

Tools perform operations, resources expose readable context, and prompts provide reusable templates. Choosing the correct primitive makes an MCP server easier to understand and govern.

`crabbox attach` follows the recorded events of an active coordinator run and

`crabbox azure` groups Azure provider setup commands. It currently has a single

`crabbox bench` records and reports local benchmark timing observations. It is a

`crabbox cache` inspects, purges, or warms package and build caches on a

`crabbox capsule` captures, replays, and tracks lightweight failure capsules.

Save the state of a lease, then restore it onto another box or fork it into a

`crabbox claims list` prints the lease claims stored on the current machine. It

`crabbox cleanup` sweeps direct-provider machines and local provider state that

`crabbox code` bridges a Linux lease's `code-server` workspace into the

`crabbox config` inspects and updates user configuration. It has three

`crabbox connect` resolves a lease and opens an interactive SSH session to it.

`crabbox cp` copies files or directories between the host and a Crabbox-owned

`crabbox desktop` drives a visible desktop session on a lease that was warmed

`crabbox doctor` runs a preflight before you commit to a long workflow. It is

`crabbox egress` gives a lease mediated outbound network: a lease-local browser

`crabbox events` prints the broker's event log for a recorded run.

`crabbox heartbeat` refreshes the idle deadline for one owned lease and prints

`crabbox history` lists recorded remote command runs from the broker. Each run is

`crabbox image` holds the trusted-operator controls for provider base images:

`crabbox init` onboards the current repository: it writes the minimal config

Your car as a chat-room agent: Raspberry Pi 5 + dashcam + local AI. CodeWatch's sibling for the garage.

The enterprise gateway for autonomous agents. Identity management, per-channel isolation, credential vault, per-session tamper-evident audit log.

AI agent for Home Assistant — talk to your home, create automations in plain language, analyze cameras with face recognition, and get proactive anomaly alerts.…

A desktop app to prototype agent ideas, inspect every harness step, replay failures, and evaluate performance, all in one place. Local-first, cloud-ready for ma…

Local-first AI memory you can see, edit, and override — portable across Claude Code, Codex, Cursor, Windsurf, and other MCP coding tools.

Web & desktop (Electron) client for the OCTO open workplace — one React + TypeScript codebase shipping browser and PC surfaces, with first-class AI agent UX.

同花顺官方 A股金融数据服务，提供股票实时行情、历史行情、财务报表、指数、板块、涨停等数据，适用于 AI Agent、量化研究和应用开发，支持 API、MCP、CLI 和 Python。Official Tonghuashun (HiThink) A-share financial data service provi…

Open-source desktop SQL workspace for PostgreSQL, MySQL/MariaDB, SQLite and 15+ more databases like DuckDB, ClickHouse, Redis and Firestore. Built-in MCP server…

🤖 Local-assist BOSS Zhipin CLI for AI agents — search, welfare filtering, shortlist, JSON-envelope output; low-risk & compliant by default.

The local-first sidebar AI agent for ComfyUI — runs on your own Claude OR ChatGPT subscription (no API keys, no extra LLM costs). Drives your live graph: edits,…

Persistent memory for Claude Code — identity, context, and continuity across sessions

🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库，覆盖8大社会科学学科的实证研究。CoPaper.AI…

A curated list of plugins, skills, MCP servers, patch/profile layers, orchestrators & UIs for DeepSeek Harness (DSH). Visualization · PPT · Coding · Agents · Lo…

World Memory Protocol.

Run your own organization of agents.

The most RAM efficient harness

AI-native, free, open-source alternative to Jira, Trello, ClickUp & Monday. Built for Scrum teams where humans and AI agents collaborate as equals — on the same…

Free open-source desktop app that scrapes JAV metadata and generates NFO + cover art for Jellyfin, Emby & Kodi. No Docker, no CLI — one-click install on Windows…

A cross-platform desktop app to manage Agent Skills in one place and sync them to multiple AI coding tools’ global skills directories — “Install once, sync ever…

天枢 (Tianshu) 是一个基于harness工程的终端编程智能体运行时（Tui X Gui），针对DeepSeek V4 做了前缀缓存工程优化（长会话实测稳态命中率 97–99%）和深度适配。它跳出了传统 AI 编程助手把大模型仅当成“工具”的局限，基于认知虚拟机 (CVM)、自感知层和信息素（Stigmergy…