AgentSearch

Self-hosted search API for AI agents. 17 endpoints. Kill-chain content extraction with optional browser rendering. Optional Tor-anonymized stack. No third-party search API keys, no per-query fees, no vendor lock-in. Optional local bearer auth is supported.

git clone https://github.com/brcrusoe72/agent-search.git
cd agent-search
./scripts/prepare-searxng.sh
docker compose up -d
curl "http://localhost:3939/search?q=distributed+consensus+algorithms"

You now have a deduplicated, multi-engine search API running on :3939.

If you enable auth, pass the token on all non-health endpoints:

export AGENT_SEARCH_TOKEN="change-me"
curl -H "Authorization: Bearer $AGENT_SEARCH_TOKEN" \
  "http://localhost:3939/search?q=distributed+consensus+algorithms"

Prefer not to use Docker for the API server?

git clone https://github.com/brcrusoe72/agent-search.git
cd agent-search
./scripts/install-native.sh
./scripts/run-native.sh

Native mode requires Python 3.11+ and a reachable SearXNG instance with JSON output enabled. It stores AgentSearch state in ./data. See Native Install.

Verify

Run the self-contained test suite:

python -m venv .venv
. .venv/bin/activate
pip install -r requirements.txt pytest requests
pip install -e sdk -e mcp-server
./scripts/prepare-searxng.sh
pytest tests -q
python -m compileall app adapters mcp-server/agent_search_mcp scripts sdk -q
docker compose -f docker-compose.yml config --quiet
docker compose -f docker-compose.yml -f examples/compose.private.yml config --quiet
docker build -t agent-search-api:test .

Those tests mock SearXNG, so they do not require Docker or a running local service.

Run the optional live localhost check:

AGENTSEARCH_INTEGRATION=1 pytest tests -q

If your local instance requires auth:

AGENT_SEARCH_TOKEN="change-me" AGENTSEARCH_INTEGRATION=1 pytest tests -q

Run the optional Docker smoke tests against a running direct/private stack:

AGENT_SEARCH_TOKEN="change-me" \
AGENTSEARCH_DOCKER_INTEGRATION=1 \
pytest tests/test_live_docker.py -q

What it does

AgentSearch wraps SearXNG with a FastAPI layer that adds everything LLM agents actually need: deduplication, cross-engine scoring, content extraction, query expansion, domain trust scoring, prompt injection scrubbing, and self-improvement.

Standard stack — docker compose up gives you search on :3939.

Private stack — docker compose -f docker-compose.yml -f examples/compose.private.yml up adds an anonymized instance on :3940 that routes all traffic through Tor with Snowflake obfuscation. Encrypted DNS via CoreDNS → Cloudflare DoT. Network-level isolation — the private stack physically cannot egress without Tor.

Search engines

AgentSearch delegates engine support to the connected SearXNG instance. The authoritative list for a running stack is:

curl "http://localhost:3939/engines"

The bundled searxng/settings.example.yml explicitly enables 25 engines, including best-effort Google/Startpage/Yahoo entries plus Brave, Bing, DuckDuckGo, Google Scholar, Semantic Scholar, arXiv, Crossref, OpenAlex, PubMed, Bing News, Reuters, Wikinews, Wikipedia, Wikidata, Hugging Face, Reddit, Hacker News, Stack Overflow, GitHub, Docker Hub, and Lobsters.

Run ./scripts/prepare-searxng.sh to create ignored local runtime files at searxng/settings.yml and searxng/settings.tor.yml with generated SearXNG instance secrets. Do not commit those generated files.

Because SearXNG is configured with use_default_settings: true, your live instance may expose additional enabled engines from the installed SearXNG catalog. Use the engines= query parameter to request specific engines, and use /engines to verify what is available in that deployment.

Strategy modes can also use direct no-key providers for vertical search. These do not require paid search APIs: GitHub repository search, MDN search, Docker Hub search, PyPI package metadata, Wikipedia, Wikidata, Hacker News, arXiv, Crossref, OpenAlex, and Semantic Scholar are called directly when a mode selects them. SearXNG remains the broad-web provider for Google/Bing/Brave/DuckDuckGo-style engines, but Google, Startpage, Yahoo, and Reddit are best-effort explicit sources rather than defaults because they are commonly blocked or empty.

Why not just use SearXNG directly?

SearXNG finds pages. AgentSearch finds pages, reads them, scores them, deduplicates them, caches them, scrubs prompt injections out of them, detects paywalls, falls back through escalating extraction strategies when the first one fails, and gets better at it over time. One API call.

	AgentSearch	Tavily	Exa	SerpAPI	Raw SearXNG
Cost	Free	$0.005/query	$0.001/query	$50/mo	Free
Self-hosted	✅	❌	❌	❌	✅
Content extraction	Kill chain + browser renderer	Basic	Basic	❌	❌
Deduplication	Cross-engine	❌	❌	❌	❌
Prompt injection scrubbing	✅	❌	❌	❌	❌
Self-improving	✅ (evolver)	❌	❌	❌	❌
Tor anonymization	Optional	❌	❌	❌	Manual

Endpoints

Search

Endpoint	Method	What it does
`/search`	GET	Multi-engine web search with deduplication and scoring
`/search/strategy`	GET	Named search modes: general, code, academic, news, private, reference, community
`/search/deep`	GET	Server-side query expansion — runs variations in parallel, fuses results
`/search/extract`	GET	Search + inline content extraction in one call
`/search/jobs`	GET	Job search across LinkedIn, Indeed, Glassdoor, ZipRecruiter
`/search/policy`	GET	Policy and regulatory document search
`/search/sources`	GET	Source discovery with institutional filtering
`/search/sources/institutions`	GET	List source registry institutions
`/search/stats`	GET	Query statistics and cache metrics
`/news`	GET	Structured multi-source news with reliable defaults and explicit engine overrides

Content extraction

Endpoint	Method	What it does
`/read`	GET	Kill chain extraction for any URL
`/read/batch`	POST	Concurrent multi-URL extraction in one request
`/providers/browser/fetch`	GET	Ephemeral browser render/extract for JS-rendered target pages

The kill chain escalates through strategies until one succeeds:

Direct fetch + smart content selectors
Readability scoring (paragraph density vs link density)
User-agent rotation (Chrome/Safari/Firefox/Edge signatures)
Browser render/extract for JS-rendered target pages
Wayback Machine (CDX API → latest snapshot)
Google Cache
Search-about fallback (find coverage elsewhere)
Custom adapters (pluggable Python modules from disk)
PDF extraction (pdfplumber)
YouTube transcript (yt-dlp)

Every request gets SSRF protection, prompt injection detection, paywall detection, and content length caps automatically. The browser renderer uses an ephemeral context, blocks high-cost resource types by default, and reports CAPTCHA/challenge pages instead of trying to bypass them. It is for rendering target pages, not for scraping blocked search-result pages.

Self-improvement

Endpoint	Method	What it does
`/adapt/report`	POST	Report a fetch failure for a URL
`/adapt/stats`	GET	View adaptation metrics and failure patterns
`/adapt/evolve`	POST	Trigger self-improvement cycle — analyzes failures, tunes config

Infrastructure

Endpoint	Method	What it does
`/health`	GET	Health check (API + SearXNG status)
`/engines`	GET	List available search engines and their status
`/providers/health`	GET	Summarize provider health from recorded live attempts
`/providers/stats`	GET	Rolling provider/SearXNG attempt telemetry

Quick examples

Search with content extraction

curl "http://localhost:3939/search/extract?q=python+async+patterns&count=3"

Returns search results with extracted content inline — no second round-trip to /read.

Strategy search

curl "http://localhost:3939/search/strategy?q=fetch+api&mode=code&count=5"
curl "http://localhost:3939/search?q=AI+regulation&mode=academic&count=5"
curl "http://localhost:3939/search?q=Python&mode=reference&count=5"

Modes validate or call only their declared sources instead of falling back silently: general tries Bing first, then uses DuckDuckGo/Brave and direct reference/community providers only when more coverage is needed; code uses direct GitHub, MDN, Docker Hub, and PyPI providers; academic uses direct arXiv, Crossref, OpenAlex, and Semantic Scholar providers; news uses Reuters, Bing News, DuckDuckGo News, and Wikinews through SearXNG; reference uses direct Wikipedia and Wikidata providers; community uses direct Hacker News; private avoids broad general web engines.

Each strategy response includes meta.engine_attempts with source/provider, query, raw result count, latency, and upstream errors so blocked or empty providers stay visible.

Provider telemetry is available without running another probe:

curl "http://localhost:3939/providers/health"
curl "http://localhost:3939/providers/stats"

Telemetry is in-memory and reflects live attempts since the API process started. It tracks attempts, successes, empty-result rate, errors, latency, last error, and last success per direct provider or SearXNG pack.

Deep search (query expansion)

curl "http://localhost:3939/search/deep?q=ethon+industrial+ai+platform&count=10"

Server-side query variation + parallel execution + result fusion. Surfaces results that flat /search misses.

Read a URL (kill chain)

curl "http://localhost:3939/read?url=https://example.com/paywalled-article"

{
  "url": "https://example.com/paywalled-article",
  "content": "Full article text extracted via Wayback Machine...",
  "strategy": "wayback",
  "chars": 4821,
  "cached": false,
  "strategies_tried": ["direct", "readability", "ua_rotation", "wayback"]
}

Browser render a JS page

curl "http://localhost:3939/providers/browser/fetch?url=https://example.com/app&max_links=20"

Returns rendered text, page title, final URL, extracted links, trust metadata, and challenge_detected=true if the page is a CAPTCHA or bot challenge.

Batch read

curl -X POST "http://localhost:3939/read/batch" \
  -H "Content-Type: application/json" \
  -d '{"urls": ["https://a.com", "https://b.com", "https://c.com"]}'

Python SDK

pip install agentsearch-client

from agentsearch import AgentSearch

client = AgentSearch()  # defaults to localhost:3939
results = client.search("manufacturing OEE best practices", count=5)
for r in results.results:
    print(f"{r.title} — {r.url}")

# Content extraction
page = client.read("https://example.com/article")
print(page.content[:500])

# Browser-rendered extraction for JS-heavy target pages
rendered = client.browser_fetch("https://example.com/app", max_links=20)
print(rendered.title, rendered.links[:3])

# Batch read
pages = client.read_batch(["https://a.com", "https://b.com"])
print(f"{pages.successful}/{pages.total} succeeded")

For authenticated instances, pass token=... or use AGENT_SEARCH_TOKEN, AGENTSEARCH_TOKEN, credentials/agent-search-token.txt, or ~/.config/agent-search/token.

LangChain tool

from langchain.tools import tool
import requests

@tool
def web_search(query: str) -> str:
    """Search the web using AgentSearch."""
    resp = requests.get("http://localhost:3939/search", params={"q": query, "count": 5})
    return "\n".join(
        f"- {r['title']}: {r['url']}\n  {r['snippet']}"
        for r in resp.json()["results"]
    )

MCP server (Claude Desktop, Cursor, Windsurf)

pip install mcp httpx
python mcp-server/server.py

For authenticated instances, set AGENT_SEARCH_TOKEN or run:

python mcp-server/server.py --token "change-me"

Add to Claude Desktop config:

{
  "mcpServers": {
    "agent-search": {
      "command": "python",
      "args": ["/path/to/mcp-server/server.py"]
    }
  }
}

See mcp-server/README.md for details.

Private stack (Tor + encrypted DNS)

The optional private stack adds a fully anonymized search path:

┌──────────┐    ┌──────────────┐    ┌───────────────┐    ┌──────────┐
│  :3940   │───▶│ api-private  │───▶│ searxng-priv  │───▶│   Tor    │──▶ Internet
│ (agent)  │    │ (FastAPI)    │    │ (SearXNG)     │    │(Snowflake│
└──────────┘    └──────────────┘    └───────────────┘    │ + obfs4) │
                                                          └──────────┘
                 All containers use CoreDNS → Cloudflare DoT
                 tor-internal network: no direct egress possible

What this gives you:

Your ISP sees TLS to Cloudflare (DNS) and WebRTC-looking traffic (Snowflake). Not search queries.
The private SearXNG instance lives on an internal-only Docker network with no internet route except through Tor.
Port 3939 = direct (fast), port 3940 = anonymized (slower, private).

Setup:

./scripts/prepare-searxng.sh
docker compose -f docker-compose.yml -f examples/compose.private.yml up -d --build

All private stack configs live in examples/ — copy and customize as needed.

Architecture

Port 3939 (direct)                    Port 3940 (Tor-anonymized)
     │                                      │
     ▼                                      ▼
┌─────────┐                          ┌─────────────┐
│   API   │                          │ api-private  │
│(FastAPI) │                          │  (FastAPI)   │
├─────────┤                          ├─────────────┤
│ dedup   │  ┌─────────────────┐     │ same code   │  ┌───────────────┐
│ scoring │  │    SearXNG      │     │ Tor egress  │  │ SearXNG-priv  │
│ cache   │──│ Google, Bing,   │     │ only        │──│ (tor-internal │
│ scrub   │  │ DDG, Brave,     │     └─────────────┘  │  network)     │
│ killchn │  │ /engines list   │                       └───────┬───────┘
│ trust   │  └─────────────────┘                               │
│ evolver │                                              ┌─────┴─────┐
└─────────┘                                              │    Tor    │
     │                                                   │ Snowflake │
     ▼                                                   │  + obfs4  │
┌─────────┐                                              └───────────┘
│ CoreDNS │──▶ Cloudflare DoT (encrypted DNS)
└─────────┘

Key modules (6,700 LOC)

Module	LOC	What it does
`killchain.py`	1016	Escalating content extraction and browser-render fallback
`browser_renderer.py`	320	Ephemeral browser rendering, extraction, and challenge detection
`main.py`	920	FastAPI app, 17 endpoints, auth, rate limiting
`source_tracer.py`	620	Source provenance tracking and citation chains
`scrubber.py`	539	Prompt injection detection and content sanitization
`source_library.py`	310	Curated institutional source registry
`domain_trust.py`	311	Domain trust scoring (TLD, age, reputation)
`evolver.py`	301	Self-improvement engine — failure analysis → config tuning
`content_cache.py`	241	URL-keyed content cache with TTL
`query_expansion.py`	201	Server-side query variation and fusion

Plus: 5 pluggable adapters (Cloudflare bypass, Medium, 403 handler, parse error recovery, empty content fallback), MCP server, Python SDK, test suite.

Case study: 0 → 17 frameworks per hunt

A real autonomous research agent ("the wolf") uses every AgentSearch endpoint. Before AgentSearch was wired in correctly, the agent's hand-rolled SearXNG client silently 401'd on three of four engines. Every hunt on a low-profile entity returned 0 frameworks.

After: 17 frameworks per hunt, 7/7 gaps closed. Same agent, same model, same prompts. The difference was the search infrastructure underneath.

→ Full walkthrough: case-studies/wolf.md

Configuration

Environment variables (set in docker-compose.yml or .env):

Variable	Default	Description
`SEARXNG_URL`	`http://searxng:8080`	SearXNG instance URL
`SEARXNG_IMAGE`	pinned SearXNG digest	SearXNG container image; override only when intentionally upgrading
`PYTHON_BASE_IMAGE`	pinned Python digest	API Docker base image; override only when intentionally upgrading
`COREDNS_IMAGE`	pinned CoreDNS digest	Private-stack DNS image
`SOCAT_IMAGE`	pinned socat digest	Private-stack TCP forwarder image
`TOR_BASE_IMAGE`	pinned Debian digest	Private-stack Tor proxy base image
`CACHE_TTL`	`3600`	Cache duration in seconds
`RATE_LIMIT`	`60`	Max requests per minute
`SQLITE_TIMEOUT`	`1.0`	SQLite lock wait timeout in seconds for query stats
`FETCH_LOG_RETENTION_DAYS`	`30`	Delete fetch analytics rows older than this many days during SQLite maintenance
`QUERY_LOG_RETENTION_DAYS`	`30`	Delete query analytics rows older than this many days during SQLite maintenance
`SQLITE_MAINTENANCE_INTERVAL_SECONDS`	`3600`	Interval for expired content-cache cleanup, log retention, and `PRAGMA optimize`
`SQLITE_VACUUM_MIN_DELETED_ROWS`	`1000`	Run `VACUUM` only when a maintenance pass deletes at least this many rows; set `0` to disable threshold vacuum
`AGENT_SEARCH_TOKEN`	(empty)	Bearer token for auth (optional)
`ADAPTERS_DIR`	`/app/adapters`	Path to pluggable adapter modules

Limitations and security notes

Search engines, news engines, rate limits, and failure modes depend on the connected SearXNG instance. /engines is the live source of truth.
Bearer auth is a simple local API gate, not a multi-user authorization system. Treat AGENT_SEARCH_TOKEN as a shared service token.
Rate limiting is in memory. It resets on restart and is per API process.
Query statistics use local SQLite with WAL and a bounded lock timeout. For high-volume multi-worker deployments, move query logging to an external database or telemetry backend.
The MCP package intentionally bounds its mcp dependency to the tested 1.27.x line. Upgrade deliberately and run the package/CI checks before publishing.
Content extraction validates the starting URL and every redirect hop before fetching redirected content, but fetched third-party pages are still untrusted and are scrubbed before being returned.
Google Cache is unreliable because public cache availability changes frequently.
The Tor/private stack is intentionally slower than direct search.

Development

pip install -r requirements.txt
SEARXNG_URL=http://localhost:8080 uvicorn app.main:app --reload --port 3939

# Run tests
pytest tests/

Release and GitHub governance

Test runs on every push and pull request.
CodeQL runs on push, pull request, and a weekly schedule.
Dependabot watches Python packages, Dockerfiles, and GitHub Actions.
Version releases are created from semantic tags such as v2.0.1 or by manually running the Release workflow with a tag input.
Update CHANGELOG.md before creating a release tag.

Contributing

Fork → branch → commit → PR.

Issues and PRs welcome. If you're building an agent that needs search, this is for you.

License

The root AgentSearch API, SDK, Docker stack, and docs are MIT licensed. The MCP server under mcp-server/ is AGPL-3.0 licensed; see mcp-server/LICENSE.

Name		Name	Last commit message	Last commit date
Latest commit History 63 Commits
.github		.github
adapters		adapters
app		app
case-studies		case-studies
credentials		credentials
docs		docs
examples		examples
mcp-server		mcp-server
scripts		scripts
sdk		sdk
searxng		searxng
tests		tests
.env.example		.env.example
.gitignore		.gitignore
CHANGELOG.md		CHANGELOG.md
Dockerfile		Dockerfile
LICENSE		LICENSE
README.md		README.md
UPGRADE-PLAN.md		UPGRADE-PLAN.md
docker-compose.yml		docker-compose.yml
pyproject.toml		pyproject.toml
requirements.txt		requirements.txt

Folders and files

Latest commit

History

Repository files navigation

AgentSearch

Verify

What it does

Search engines

Why not just use SearXNG directly?

Endpoints

Search

Content extraction

Self-improvement

Infrastructure

Quick examples

Search with content extraction

Strategy search

Deep search (query expansion)

Read a URL (kill chain)

Browser render a JS page

Batch read

Python SDK

LangChain tool

MCP server (Claude Desktop, Cursor, Windsurf)

Private stack (Tor + encrypted DNS)

Architecture

Key modules (6,700 LOC)

Case study: 0 → 17 frameworks per hunt

Configuration

Limitations and security notes

Development

Release and GitHub governance

Contributing

License

About

Topics

Resources

License

Uh oh!

Stars

Watchers

Forks

Releases

Packages 0

Uh oh!

Uh oh!

Contributors

Uh oh!

Languages

Packages