# ๐Ÿ”ง MCP Tool Stack โ€” 13 servers, ~180 tools, AI-native dev workflow > Source: > Published: 2026-05-22 09:40:47+00:00 # ๐Ÿ”ง MCP Tool Stack ยท Enes Demir > **Model Context Protocol** โ€” 13 servers, ~180 tools, AI-native development > > _Last updated: 2026-05-22 ยท Hermes Agent ยท DeepSeek V4 Flash_ --- ## ๐Ÿ“ก The Stack | # | Server | Tools | Area | Link | |---|--------|-------|------|------| | 1 | **Firecrawl** | ~24 | Web scraping, search, crawl, extraction, monitoring | [firecrawl.dev](https://firecrawl.dev) | | 2 | **GitHub** | ~30+ | PRs, issues, repos, code review, CI, releases | [MCP GitHub Server](https://github.com/modelcontextprotocol/servers/tree/main/src/github) | | 3 | **HuggingFace** | ~20 | Hub search, Spaces, papers, datasets, HF Jobs | [huggingface.co/mcp](https://huggingface.co/docs/hub/en/mcp) | | 4 | **Kaggle** | ~50+ | Competitions, datasets, models, notebooks, leaderboards | [kaggle.com/docs/mcp](https://www.kaggle.com/docs/mcp) | | 5 | **W&B (Weights & Biases)** | ~15 | Run tracking, traces, artifacts, reports, evals | [github.com/wandb/wandb-mcp-server](https://github.com/wandb/wandb-mcp-server) | | 6 | **arXiv** | ~12 | Paper search, download, citation graph, alerts | [github.com/blazickjp/arxiv-mcp-server](https://github.com/blazickjp/arxiv-mcp-server) | | 7 | **Browser Use** | ~13 | Headless browser automation, screenshots, interaction | [github.com/browser-use/browser-use](https://github.com/browser-use/browser-use) | | 8 | **Filesystem** | ~13 | File read/write, search, directory tree, metadata | [MCP Filesystem Server](https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem) | | 9 | **Context7** | ~2 | Library documentation query (Next.js, React, etc.) | [mcp.context7.com](https://mcp.context7.com) | | 10 | **Git** | ~10 | Local git operations (diff, log, branch, commit) | [mcp-server-git](https://github.com/modelcontextprotocol/servers/tree/main/src/git) | | 11 | **Time** | ~2 | Timezone conversion, current time | [MCP Time Server](https://github.com/modelcontextprotocol/servers/tree/main/src/time) | | 12 | **Sequential Thinking** | ~1 | Structured multi-step reasoning | [@modelcontextprotocol/server-sequential-thinking](https://github.com/modelcontextprotocol/servers/tree/main/src/sequentialthinking) | | 13 | **Supabase** | ~? | Database, auth, storage via MCP | [supabase.com/docs/guides/platform/mcp](https://supabase.com/docs/guides/platform/mcp) | --- ### ๐Ÿ”ฅ Firecrawl ``` crawl โ†’ scrape โ†’ search โ†’ map โ†’ extract โ†’ agent ``` [๐ŸŒ firecrawl.dev](https://firecrawl.dev) โ€” Full-stack web intelligence: - **scrape** โ€” clean markdown from any URL, JSON schema extraction - **search** โ€” web search with snippet scraping - **crawl** โ€” multi-page extraction with depth control - **map** โ€” discover URLs on a domain - **agent** โ€” autonomous research across unknown pages - **interact** โ€” click, fill forms, navigate after scrape - **monitor** โ€” recurring diffs (markdown or JSON path tracking) - **parse** โ€” local PDF/Word/Excel โ†’ markdown > **Use case**: Documentation extraction, competitive research, content monitoring, data pipeline ingestion. --- ### ๐Ÿ™ GitHub ``` repos โ†’ issues โ†’ PRs โ†’ code review โ†’ actions โ†’ releases ``` [๐Ÿ™ MCP GitHub Server](https://github.com/modelcontextprotocol/servers/tree/main/src/github) โ€” Full GitHub lifecycle: - Repositories: clone, create, fork, manage remotes - Pull requests: diff review, inline comments, merge - Issues: create, triage, label, assign, search - Actions: workflow status, logs - Releases, gists, discussions, projects - Multi-toolset architecture: granular permission scoping > **Use case**: PR code review automation, issue triage, release management. --- ### ๐Ÿค— HuggingFace ``` hub โ†’ spaces โ†’ papers โ†’ datasets โ†’ jobs ``` [๐Ÿค— huggingface.co/mcp](https://huggingface.co/docs/hub/en/mcp) โ€” AI/ML ecosystem: - **Hub search** โ€” models, datasets, Spaces by query/filter - **Repo details** โ€” overview, dataset structure, preview rows - **Space search** โ€” semantic search, MCP-enabled Spaces, dynamic invocation - **Paper search** โ€” ML research papers on HF hub - **HF Jobs** โ€” remote compute (Python/UV/Docker) for batch inference, evaluation - **Doc search** โ€” HF library documentation - **Duplicate Space** โ€” copy Spaces (free CPU/GPU tier) > **Use case**: Model discovery, dataset exploration, batch inference on remote compute. --- ### ๐Ÿ† Kaggle ``` competitions โ†’ datasets โ†’ models โ†’ notebooks โ†’ leaderboards ``` [๐Ÿ† kaggle.com/docs/mcp](https://www.kaggle.com/docs/mcp) โ€” Data science & ML competitions: - **Competitions** โ€” search, metadata, data files, submissions, leaderboards - **Datasets** โ€” search, info, file listing, metadata update, upload - **Models** โ€” list, create, update, variations, version files - **Notebooks** โ€” create session, save+run, output download, status - **Forums** โ€” topics, comments, write-ups, hackathon tracks - **Benchmarks** โ€” create tasks, leaderboards > **Use case**: Competition analysis, dataset ingestion, model versioning, write-up research. --- ### ๐Ÿ“Š W&B (Weights & Biases) ``` runs โ†’ artifacts โ†’ traces โ†’ reports โ†’ registry ``` [๐Ÿ“Š github.com/wandb/wandb-mcp-server](https://github.com/wandb/wandb-mcp-server) โ€” ML experiment & LLM observability: - **Runs**: query, filter, sort, history sampling, diagnosis - **Artifacts**: list versions, compare diffs, lineage - **Traces** (Weave): LLM call tracing, eval summary, schema inference - **Reports**: create markdown reports with native chart panels - **Registry**: list entities, registries, collections, versions - **Analysis**: log computed data, diagnose run health > **Use case**: Experiment tracking, LLM observability, eval dashboards, model registry. --- ### ๐Ÿ“„ arXiv ``` search โ†’ abstract โ†’ download โ†’ read โ†’ citation graph โ†’ alerts ``` [๐Ÿ“„ github.com/blazickjp/arxiv-mcp-server](https://github.com/blazickjp/arxiv-mcp-server) โ€” Academic research: - **Search** โ€” advanced query syntax, field filters, category scoping - **Abstract** โ€” quick relevance check before download - **Download** โ†’ **Read** โ€” full paper in markdown - **Citation graph** โ€” Semantic Scholar references/citations - **Semantic search** โ€” similarity search across downloaded library - **Watch topics** โ€” standing alerts for new papers > **Use case**: Literature review, citation tracking, research monitoring. --- ### ๐ŸŒ Browser Use ``` navigate โ†’ click โ†’ type โ†’ extract โ†’ screenshot โ†’ scroll ``` [๐ŸŒ github.com/browser-use/browser-use](https://github.com/browser-use/browser-use) โ€” Headless browser with full interaction: - Multi-tab management (list, switch, close tabs) - Element interaction via index or coordinates - HTML extraction, content query, screenshots - Session management (list, close sessions) - Agent mode for complex multi-step tasks > **Use case**: Dynamic SPA content, login-required pages, form automation. --- ### ๐Ÿ“ Filesystem ``` read โ†’ write โ†’ search โ†’ tree โ†’ edit โ†’ move ``` [๐Ÿ“ MCP Filesystem Server](https://github.com/modelcontextprotocol/servers/tree/main/src/filesystem) โ€” Local file operations: - Read/write files, directory listing with sizes - Recursive tree view, file search with glob patterns - Line-based edit with diff output, file move/rename - Media file read (images, audio as base64) > **Use case**: Code analysis, file manipulation, project exploration. --- ### ๐Ÿ“š Context7 ``` resolve library โ†’ query docs ``` [๐Ÿ“š mcp.context7.com](https://mcp.context7.com) โ€” Up-to-date library documentation: - **Resolve Library ID** โ€” map package name to Context7 ID - **Query Docs** โ€” retrieve current docs + code examples with versioning > **Use case**: Framework documentation queries (Next.js, React, Tailwind, etc.). --- ### ๐Ÿ”— Git ``` status โ†’ diff โ†’ log โ†’ branch โ†’ commit โ†’ add ``` [๐Ÿ”— mcp-server-git](https://github.com/modelcontextprotocol/servers/tree/main/src/git) โ€” Local git operations: - Status, staged/unstaged diffs, commit logs - Branch creation, checkout, listing - Commit with message, add files, reset staged - Diff between branches/commits > **Use case**: Local version control within agent workflow. --- ### ๐Ÿ• Time ``` current time โ†’ convert time ``` [๐Ÿ• MCP Time Server](https://github.com/modelcontextprotocol/servers/tree/main/src/time) โ€” Timezone utilities: - Get current time in any IANA timezone - Convert between timezones --- ### ๐Ÿง  Sequential Thinking ``` structured reasoning ``` [๐Ÿง  @modelcontextprotocol/server-sequential-thinking](https://github.com/modelcontextprotocol/servers/tree/main/src/sequentialthinking) โ€” Meta-cognitive reasoning tool: - Dynamic, reflective problem-solving - Branch, revise, backtrack through thoughts - Hypothesis generation and verification > **Use case**: Complex problem decomposition, debugging, planning. --- ### โšก Supabase ``` database โ†’ auth โ†’ storage โ†’ realtime ``` [โšก supabase.com/docs/guides/platform/mcp](https://supabase.com/docs/guides/platform/mcp) โ€” Backend-as-a-service via MCP: - PostgreSQL queries, table management - Authentication, user management - File storage - Realtime subscriptions --- ## ๐Ÿ“Š By the Numbers ``` Total MCP Servers : 13 Total Tools : ~180 Primary Agent : DeepSeek V4 Flash Host : macOS (Apple Silicon) CLI : Hermes Agent โ†’ OpenCode Go ``` --- ## ๐Ÿง  Why This Stack This isn't random โ€” each server fills a role: ``` Research : arXiv + HuggingFace Papers + Firecrawl Search Development : GitHub + Git + Filesystem + Context7 Data/ML : Kaggle + HuggingFace Hub + W&B Automation : Browser Use + Firecrawl Agent + HF Jobs Reasoning : Sequential Thinking + Time Backend : Supabase ``` --- > **Tip**: Stack evolves. This is a snapshot; new servers get added as workflows demand. > > _Part of my dev workflow โ€” Enes Demir_