Hound Web MCP

🐕 Hound

Give your AI agent the web. $0. Two commands. No keys.

Fetch · crawl · bypass bot walls · read PDFs (even scanned) · search the web One MCP server · one warm browser · zero accounts · runs on your machine

pip install hound-mcp[all] && playwright install chromium

Install · The 6 tools · Search · Comparison · Honest limits

🎬 Demo

Same prompt, three tools. Hound does the whole thing on its own, search + fetch + crawl, locally. The others get stuck on the parts they don't do.

✨ New in 12.1.2

BYOK search · intent-aware multi-query fan-out · six-signal ranking · next-gen stealth engine.

• 🔑 Bring Your Own Key (BYOK) search: add your own API keys for Serper, Tavily, Exa, Firecrawl, or TinyFish via hound keys add. Keys become the primary search source with key stacking (multiple keys per provider, auto-rotation on rate limit), automatic fallback to Hound's keyless local engines when all keys are exhausted. CLI key management: hound keys add/list/test/remove/clear.
• 🧠 Intent-aware multi-query fan-out: Hound detects query intent (comparison, howto, research, code, reference, news, factual) and generates expanded query variants distributed across diversity engines. Same parallel request count, zero added latency, wider recall. Cross-variant consensus: a URL surfaced by different queries from different engines is a stronger authority signal.
• 📊 Six-signal ranking: cross-variant consensus + domain reputation + answer-signal scoring + title relevance + URL relevance + result diversity (max 2 per domain in top results). Source-type detection tags every result as docs/paper/repo/blog/forum/reference/news.
• 🧬 Next-gen stealth engine: system Chrome auto-detection, 4 coherent fingerprint profiles, JS-layer patches (HeadlessChrome UA fix, navigator.webdriver=undefined, canvas noise, permissions API), human behavior simulation (Bezier mouse curves, natural scroll), CF Turnstile solver with human-like mouse movement. Passes bot.sannysoft.com, bypasses Cloudflare Turnstile on CanadianInsider (hardest in a 31-site benchmark), Medium, StackOverflow, NowSecure, Glassdoor (DataDome). See the stealth benchmark below.
• 🐍 No more scrapling. All scrapling functionality replaced with hound's own modules: fetcher.py (primp-based HTTP), browser.py (patchright-based browser), extractor.py (trafilatura + markdownify). Smaller install, fewer transitive deps, faster cold start.
• 📱 Works everywhere. Lean install pip install hound-mcp pulls no browser deps. On platforms without playwright (Termux, aarch64), hound runs in HTTP-only mode with graceful degradation.
• 🔒 Reliability fixes: universal error detection (error pages no longer look like success), dead Internet Archive fallback removed, self-healing CLI + stale process cleanup, CSS selector errors propagate instead of silently returning [], browser startup failure cleans partial sessions, hound --rollback works for pinned older versions, focus and actions now forwarded from MCP dispatcher to smart_fetch.
• 🐳 Docker support: multi-stage Dockerfile, docker-compose with shm_size 1gb, non-root user, healthcheck. By @imonlinux.
• 🧪 673 tests.

Why you should pick Hound

Hound is one MCP server that gives any agent (Claude Code, Cursor, OpenCode, Hermes, Pi, anything that speaks MCP) full web research from a single local process.

• 🆓 $0 forever, MIT: no keys, no accounts, no per-request billing, no data routed to a third-party scraper. Search is keyless and local.
• 🧠 Mastered on connect: a one-time instructions block hands the agent the mental model, the #1 workflow, and the known limits. Effective on turn one.
• 📐 ~2.9K tokens, 6 tools: hand-crafted tool defs, no Pydantic schema bloat. More capability than tools shipping 5K+.
• 🎯 Every response is actionable: content_ok, next_action, summary, page_type, content_age_days/is_stale, source_type/is_official, relevance_score, fetch_relevance. Agents branch on structured fields, not error text. Hard-blocks (404/bot/auth) return clean errors, not fake content.
• 🛡️ Production-safe startup + shutdown: cold start under 1s so the MCP handshake never times out; exits 0 with clean stderr, no crash-like teardown noise.

Hound is for the agent itself. You install it once; the agent calls it whenever it needs the web.

🚀 Quick start

pip install hound-mcp[all]          # fetch + crawl + keyless search + PDF + OCR + neural rerank
playwright install chromium         # the anti-detect browser engine

Then point any MCP client at the hound command. No arguments, no keys, no env vars. See Install for the lean option and Tell your agent to install it for a copy-paste prompt.

hound -v          # version + update status
hound -u          # update to latest (brick-proof, self-healing)
hound --doctor    # health check + fix advice
hound --rollback  # undo the last update

If hound ever breaks (a failed update, a locked launcher), recover with python ~/.hound/repair.py, or run hound --doctor to diagnose.

🐳 Docker

Hound can run in Docker with HTTP mode, ideal for:

• Running on a server
• Isolated deployment

Quick start with Docker Compose

git clone https://github.com/dondai1234/master-fetch.git
cd master-fetch
docker-compose up -d

Hound will be available at http://<your-host-ip>:8765/mcp as an HTTP MCP server (use localhost if running on the same machine).

Manual Docker build

docker build -t hound-mcp .
docker run -p 8765:8765 hound-mcp

Environment variables

Connecting to HTTP mode

For MCP clients that support HTTP (Claude Code, Open WebUI), use your host's LAN IP:

http://<your-host-ip>:8765/mcp

(Or localhost:8765/mcp if running on the same machine.)

🧰 The 6 tools

🔎 Local keyless search

No API key, no account, no third-party service. smart_search runs 10 keyless backends in parallel on your machine, merges, dedups, and ranks. It returns URLs + ranking, not page content: the agent smart_fetches whichever results match what it needs (the ranking is a hint, not a directive).

• 🌐 10 independent backends: duckduckgo, brave, mojeek, yahoo, yandex, startpage, google, qwant, plus opt-in wikipedia + grokipedia. Six+ independent index families, not the same feed twice.
• 🧠 Neural rerank: a local ONNX cross-encoder (ms-marco-MiniLM-L-6-v2, Apache-2.0) running on the onnxruntime Hound already ships for OCR. Exa-style semantic ranking, $0, on your machine. Model downloads once (~80MB, cached, not bundled). Lean installs fall back to cross-engine consensus + engine-position order.
• 🎯 Cross-backend consensus: a URL returned by several independent indexes gets a consensus boost: a free authority signal from merging, no extra fetches. Every result carries relevance_score (0–1), fetch_relevance (high/med/low), and engines_consensus.
• 🔍 find_similar: pass url=; Hound fetches a page you like, derives a query, and reranks candidates against that source page. Exa's find-similar, local.
• 🛡️ Never dead: a diversity quorum waits for at least 3 backends to contribute before returning, so a single backend's bias or rate-limit can't dominate. A backend that CAPTCHAs or rate-limits is circuit-broken for 60s and carried by the others. engine_blocked in the response reports which ones cooled down.
• 📊 Filters: site / exclude_sites (domain include/exclusion), location / language / region (geo), page (0–10), freshness (day | week | month | year). Default 6 results. A quality filter drops low-relevance results instead of padding to the max with garbage.
• 📈 related_queries: follow-up queries mined from result titles + snippets (no LLM). Search one to refine a broad query.

Search is 100% HTTP: it never touches the browser (the single Patchright browser is smart_fetch's alone).

🔑 Bring Your Own Key (BYOK)

Hound's local keyless search works with zero configuration and never goes away. But some users have API keys from search providers, either from free tiers or paid plans. Hound respects that: bring your keys, and Hound makes them first-class citizens with the same reliability guarantees as its own keyless engines.

When keys are configured, those providers become the primary search source and Hound's local keyless engines are completely shut off. This is the entire point of BYOK: avoid hitting public search engines from your IP. The local engines run only as a last-resort fallback when every API key is exhausted or rate-limited, so the search never fails.

Hound uses one provider per search. If you have keys for multiple providers, Hound picks the first available one, and only switches to the next provider when the first is exhausted. This means your search capacity scales with how many keys you configure for a single provider, not with how many providers you stack.

Supported providers

Mix and match. Use one provider, use all five, switch providers anytime. Hound treats them as parallel engines in the same ranking pipeline.

Key stacking

Add multiple keys per provider. Hound stacks them in a rotation pool per provider:

• 🔁 Auto-rotation: when a key hits a rate limit (HTTP 429), Hound switches to the next key for the same provider within milliseconds. The search completes without the agent ever knowing a key was rate-limited.
• ⏳ Per-key cooldown: a rate-limited key enters a 60-second cooldown. An invalid key (401/403) enters a 300-second cooldown. Hound keeps trying the remaining keys in the pool.
• 🛡️ Graceful exhaustion: only when all keys for all providers are exhausted does Hound fall back to its keyless local engines. The transition is seamless: the agent gets results either way.
• 📊 Live key testing: hound keys test makes a real API call per key and reports which are valid, rate-limited, or invalid. Know before you search.

This means your search capacity scales with how many keys you configure, not with a single key's rate limit.

CLI key management

# Add a key (stack multiple keys for the same provider)
hound keys add serper YOUR_SERPER_KEY
hound keys add serper ANOTHER_SERPER_KEY   # stacked, auto-rotated
hound keys add tavily YOUR_TAVILY_KEY

# List all configured keys (redacted for safety)
hound keys list

# Test all keys (live API call per key)
hound keys test

# Test a specific provider
hound keys test serper

# Remove a specific key by index (0-based)
hound keys remove serper 0

# Remove all keys for a provider
hound keys remove serper

# Remove all keys across all providers
hound keys clear

Keys are stored at ~/.hound/search_keys.json with redaction on display. Key rotation state is in-memory only (resets on restart).

Environment variables

For CI/CD, Docker, or ephemeral environments, set env vars instead of the config file. Comma-separated for multiple keys:

export HOUND_SEARCH_SERPER_KEYS=key1,key2,key3
export HOUND_SEARCH_TAVILY_KEYS=key1
export HOUND_SEARCH_EXA_KEYS=key1
export HOUND_SEARCH_FIRECRAWL_KEYS=key1
export HOUND_SEARCH_TINYFISH_KEYS=key1

Env vars override the config file for any provider that has env vars set. Providers without env vars fall back to the config file. Mix both: some providers in the config file, others via env vars.

hound doctor integration

hound --doctor reports your BYOK configuration status: which providers have keys, how many keys per provider, and whether any are in cooldown. One command to see the full picture.

🌐 Fetch & anti-bot

smart_fetch tries plain HTTP first (~1s). If the site blocks HTTP or serves a JS shell, it auto-escalates to a Patchright anti-detect browser with Cloudflare challenge solving. Two tiers, nothing to configure.

• 🛡️ Built-in Cloudflare bypass: a single stealthy Chrome warms at startup. It closes after 5 min of idleness to free RAM (HOUND_BROWSER_IDLE_TIMEOUT, set 0 to keep it alive forever) and relaunches in ~2s on the next fetch. Pages close after each fetch, idle memory stays near baseline. One browser total.
• 🧬 Stealth engine (v11.1+): system Chrome auto-detection (channel=chrome for real TLS fingerprint), coherent fingerprint profiles, JS-layer patches (HeadlessChrome UA fix, navigator.webdriver=undefined, canvas noise via getImageData+toDataURL interception, permissions API), human behavior simulation (Bezier mouse curves, natural scroll, dwell time), and a Cloudflare Turnstile solver with human-like mouse movement. See the stealth benchmark below.
• 🎯 Query-focused extraction: smart_fetch(url, focus="...") returns only the BM25-relevant blocks. Cuts context 80%+ on long pages, no re-fetch (runs post-cache). Re-pass the same focus when paginating.
• 🖱️ Page interaction: actions=[{click:'button.load-more'},{fill:{selector:'#q',text:'x'}},{press:'Enter'},{wait:500},{scroll:3},{wait_selector:'.item'}] for load-more, search forms, pagination, infinite scroll. Forces stealthy + bypasses cache.
• 🏷️ Metadata on every response: title, description, site name, type, image, canonical URL, language, published time, author (OpenGraph + JSON-LD + canonical).
• 🔗 Outgoing links: include_links=true populates response.links classified as citations (main-content references, the ones worth following) / navigation / external + a primary_source hint. Follow a page's source chain in one step.
• 🐕 Reddit, optimized: Reddit URLs auto-rewrite to old.reddit.com (7× smaller) and skip to the stealthy browser. Subreddit listings parse into structured posts with promoted ads filtered out.
• 💾 Smart caching: SQLite (WAL mode), keyed by URL + extraction type + css_selector + pages. Bad content is never cached; a size cap evicts the oldest so a long-lived agent's cache can't grow unbounded. cache_ttl=0 forces fresh.
• 📐 Pagination: content over 40KB is chunked; the response gives next_offset so the agent pages through with one more call (served instantly from cache).

🧬 Stealth benchmark

Real-world results from v11.1.0, tested against hard anti-bot targets.

Detection test sites (all checks pass):

Anti-bot protected sites (content extracted):

Note: Google Search returns 429 (rate limited) as it uses its own bot detection independent of Cloudflare. This is expected.

Stealth signals verified:

Memory efficiency (5 sequential fetches):

RSS decreased by 3.5 MB over 5 fetches. No RAM creep. The Memory.simulatePressureNotification CDP command triggers Chrome's internal GC + cache drop after each fetch (~5ms, non-disruptive).

🕷️ Crawl

smart_crawl walks same-domain links in best-first order: discovered URLs are scored by focus relevance + content-likelihood (docs/guide/api boosted, login/submit/cart penalized) + shallow depth, so content pages are crawled before junk when the budget is tight.

• 🎯 Content-adaptive extraction: article/docs → trafilatura main content; list/index pages (HN, aggregators, directories) → a structured * [title](url) link list; JS shells → detected and reported honestly.
• 🗺️ Sitemap mode: options sitemap=true maps the whole site from sitemap.xml in ONE fetch (full URL list + lastmod, no BFS). sitemap='auto' uses it if the site has one, else falls back to BFS. Collapses a hundreds-of-pages discovery crawl into one call.
• 📍 discover_only=true: URL map only (BFS-based). For big sites prefer sitemap=true instead.
• 🎯 focus='query': prioritizes relevant pages within the budget AND focus-filters each page's content.
• 📋 crawl_urls=[...]: second-phase selective crawl of a chosen subset (no re-discovery).
• 🛡️ Dedup + scoping: URLs normalized so /docs and /docs/ are never crawled twice. Same-domain only by default; path_include / path_exclude to scope.
• ⏱️ Caps: max_pages (default 10), max_depth (default 2), max_total_chars (token budget), deadline_ms (overall time, default 120000). Each page carries content_ok + status + fetched_at; next_action tells you if the crawl stopped early.

📄 PDF + scanned-PDF OCR

smart_fetch detects a PDF (by content-type or %PDF magic bytes) and extracts it to structured markdown with pdfplumber (MIT): multi-column reading order, real tables as markdown tables, font-size headings, de-hyphenated paragraphs, a metadata header, and --- Page N --- markers.

• 📑 table_of_contents: the PDF outline as [{level, title, page, end_page}]. PDFs without bookmarks get a heading-based fallback map. Pass pages='23-31' to grab one section by range and save tokens.
• 🔍 CID-corruption auto-OCR (the flagship trick): academic papers embed font subsets without a Unicode map, so extractors emit (cid:71)(cid:302)... garbage for figures/diagrams/math. But the glyphs render correctly. Hound detects CID-garbage pages, renders them via pypdfium2, and OCRs them with rapidocr, recovering the real text automatically.
• 🖼️ Scanned / image-only PDFs (and image-only web pages) are auto-OCR'd too. Pure-pip, no system binary, with [all].
• 📊 quality_score (0.0–1.0) + honest content_ok: trust PDF content more the closer the score is to 1.0.
• 📎 password for encrypted PDFs; include_media=true for per-page image metadata; a .pdf URL that returns a login/paywall is reported as auth_required.

📸 Screenshot

screenshot captures a page as an image. For multimodal agents only: use when content is rendered as images / canvas / image-of-text or you need visual layout. Text-only agents should use smart_fetch instead. A stealthy browser session is auto-managed.

📊 Comparison: free tools

Most free web tools for agents do one thing and miss the rest. Hound is the only one that bolts all of it onto a single local MCP server for $0, no keys.

The short version: Crawl4AI crawls well but has no search and trips on Cloudflare. Parallel Search is remote search-only, no crawl, and runs on their servers. Jina fetches but rate-limits and routes through Jina. Firecrawl keeps the good stuff behind the paid cloud. Hound is the only free tool that combines keyless local search, built-in Cloudflare bypass, best-first crawl, scanned-PDF OCR, page interaction, and query-focused extraction in one local MIT server: $0, no accounts, no keys.

📦 Install

pip install hound-mcp[all]          # recommended: fetch + crawl + keyless search + PDF + OCR + neural rerank
playwright install chromium

pip install hound-mcp               # fetch + crawl + keyless search (HTTP-only, no stealthy browser)

hound -u          # update to latest (brick-proof: --no-deps, detached helper, self-heal)
hound --doctor    # health check: launcher, imports, metadata, deps, PyPI, repair script
hound --rollback  # reinstall the version from before the last update

python ~/.hound/repair.py   # stops hound, force-reinstalls hound-mcp from PyPI, verifies

🤖 Tell your agent to install it

Paste this into your agent:

Install the Hound MCP server on this machine. Follow every step. Do not skip any.

1. Figure out which agent harness you are running on (OpenCode, Hermes, Pi, etc). Then find: (a) where the MCP config file lives, and (b) what format it expects for adding a local MCP server. Read the harness docs if needed. Do not guess.

2. Run: pip install hound-mcp[all]
   Then run: playwright install chromium (But only if it isnt installed already, verify first about its existence)
   If either fails, stop and tell the user.

3. Find the MCP config file from step 1 and back it up before editing. Add a new MCP server named "hound" with command "hound", no arguments, in the format your harness requires. No API keys or environment variables are needed (search is keyless and local).

4. Save the file. Tell the user to restart the agent. After restart, smart_fetch, smart_crawl, smart_search, screenshot, cache_clear and version should be available.

pip install hound-mcp[all]
pi install npm:@houndmcp/hound-mcp-pi

hound -u                              # update the MCP server
pi update npm:@houndmcp/hound-mcp-pi  # update the extension

hound --http --host 127.0.0.1 --port 8765

⚠️ Honest limits

No free tool can do everything. Hound is upfront about what it can't:

When a fetch or search fails, the response says exactly why and what to try next, so the agent doesn't waste calls guessing.

🪙 Token cost

Most MCP servers cost 3–5K tokens just to exist. Hound's 6 tools cost ~2.7K tokens at tools/list (measured with cl100k_base); the connect-time instructions (~0.8K, the orientation doc) are injected ONCE at handshake, not repeated every turn. Your context window is expensive; Hound respects it.