Token-efficient, local-first web research MCP server in Rust. Keyless multi-engine search with parallel fanout, dedup and automatic failover — no API keys, no runtime, no VM required. Built on FlojoMCP over the official Rust MCP SDK (rmcp).
Token-efficient, local-first web research MCP server in Rust: keyless multi-profile metasearch. Built on FlojoMCP.
Sterne
5
Tools
—
Uptime
—
Keywords: MCP server, web search, deep research, metasearch, keyless, token efficiency, AI agents, Model Context Protocol, local-first, Rust, DuckDuckGo, Bing.
Why
Search MCPs force a bad trade: scrape fragile engines yourself, rent an API key, or run a container farm — and either way the agent pays for it in tokens and timeouts.
Keyless by default: DuckDuckGo, Bing and Brave HTML endpoints queried directly from Rust for compatibility. No accounts, no cards, no Docker, no WSL. The direct HTML set is experimental: see the measured quality and policy caveats below before treating it as a durable index.
Browser impersonation: the HTTP layer (primp, same crate ddgs uses) masquerades as Chrome153/Windows — measured difference: without it Bing serves a 302 to its homepage and Brave returns 429; with it all three engines answer 200.
Fanout with failover: all providers queried in parallel, results fused with weighted RRF and deduplicated by canonical URL. When one engine rate-limits your IP (202/403/429 happen — see below), the others still answer: research degrades, it doesn't die.
Tokens are a feature: compact results, server-side truncation budgets (src/limits.rs), dedup before the agent ever sees duplicates.
No silent failures: every search response carries the per-provider status (ok, empty, filtered, rate_limited, unreachable) and the warnings. If all engines were throttled, the tool returns a typed AllProvidersRateLimited error; if providers answered but every result failed Argos' domain/relevance gate, it returns NoUsableResults instead of pretending the topic had no hits.
Honest, typed errors: rate_limited, unreachable, all_providers_rate_limited, no_usable_results — each with an actionable hint, never a generic Error executing tool ghost.
Per-provider visibility: every search response carries { results, providers: [{name, kind}], warnings } so the agent knows exactly which engines contributed, which were empty or filtered, and which errored.
One small binary: ~7 MB exe, ~5 ms startup, ~4 MB RAM — no Python/Node runtime.
Status
Early development — Milestone 1: keyless multi-provider search. Not production-ready yet.
Feature
Status
search via five keyless profiles: general web (DuckDuckGo/Bing/Brave), academic (OpenAlex, Crossref, arXiv, Europe PMC, PubMed, DOAJ, Semantic Scholar), code (GitHub, crates.io, npm, Packagist), news (GDELT) and knowledge (Wikipedia). Weighted RRF fusion, canonical dedup, failover
search only reads: nothing the caller can observe is mutated, so a host can invoke it without warning.
Configuration
Env var
Default
Meaning
ARGOS_PROVIDERS
16 implemented adapters
Enabled adapters; the profile decides which family spends quota
ARGOS_CONTACT_EMAIL
(unset)
Optional contact address for the OpenAlex/Crossref/PubMed polite pools. Not a key, no account
ARGOS_GITHUB_TOKEN
(unset)
Optional token; raises the GitHub Search quota from 60/h
ARGOS_SEMANTIC_SCHOLAR_KEY
(unset)
Optional key; anonymous traffic is throttled hard
ARGOS_META_INITIAL
2
Maximum providers in the first native metasearch wave
ARGOS_META_TOTAL
5
Maximum providers considered across fallback waves
The rate-limit reality (measured, not guessed)
Anti-bot limits live in the upstream engines, not in your client — and every keyless approach shares that wall — SearXNG's own docs admit it gets CAPTCHAs too, which is why the optional adapter was dropped. Measurements from one residential IP,2026-09-24:
Endpoint
Plain client (reqwest, rustls and schannel)
With primp (Chrome153 impersonation)
html.duckduckgo.com
200 with browser UA / 403 with PowerShell UA
200,10-12 organic results
www.bing.com/search
302 → homepage (same headers/HTTP1.1 that curl passes — pure ClientHello discrimination)
200,10 b_algo
search.brave.com/search
429
200,20 web results
www.mojeek.com/search
CAPTCHA
CAPTCHA (IP/consent-based — out of the set)
Argos raises the ceiling with design instead of hoping: browser impersonation at the transport layer, parallel fanout across independent rate budgets, weighted RRF fusion + canonical URL dedup (fewer redundant queries for the agent), a short-lived cache with single-flight, per-request timeouts and typed rate-limit errors with failover.
Quality reality (measured 2026-09-24)
HTTP 200 is not a quality result. A controlled live audit found that DDG and Brave can return highly relevant niche results on their first request and then quickly return 202/429 from the same IP. Bing returned valid HTML with b_algo entries but poor semantic coverage for the same queries: free piano VST produced free-game pages, and one Spitfire Audio LABS free run produced Macao travel pages. The query itself was present in Bing's final URL, so this was not a URL-encoding failure.
Argos therefore applies a conservative result-side quality gate:
domains is validated and enforced against the returned host; site: text is only a ranking hint, never the security boundary.
Results with no meaningful query terms are rejected as obvious entity drift.
A provider whose entire result set was rejected is filtered, not ok.
If no provider contributes a usable result, the tool returns typed NoUsableResults.
The full matrix and policy notes are in PROVIDER_AUDIT.md. The direct HTML endpoints remain a compatibility set, not a guarantee of quality or long-term provider availability.
Comparison (researched 2026-09-23)
Argos Engine
ddgs MCP (Python)
Cloud search MCPs (Tavily / Serper / Exa)
Runtime
single ~7 MB exe
Python ≥3.10 + pip package
binary + API key
Startup / RAM
~5 ms / ~4 MB
~324 ms / ~50-76 MB¹
n/a
Keys & cost
none
none (direct scraping)
keys, free tiers then paid
Multi-engine failover
✅ parallel fanout
sequential per-call engines
provider-dependent
Deadlines / typed errors
✅
❌ blocking threads, opaque errors
provider-dependent
Engine maintenance
sync from ddgs reference + own parsers
per-engine XPath, constant upstream fixes
provider's problem
Deep-research fan-out
roadmap (M2)
no — N sequential LLM calls
partial (their own APIs)
¹ FastMCP/Python figures from FlojoMCP benchmarks; ddgs behavior from source inspection of v9.16.0.
Noch keine Tool-Liste. Remote-Server erhalten eine nach dem ersten Health-Check.
FAQ
Fragen
Wie installiere ich Argos Engine?
Kopieren Sie auf dieser Seite die stdio-Konfiguration für Cursor, Claude Desktop, Claude Code oder VS Code. Sie nutzt die Installation des Herausgebers, keinen PlugMCP-Proxy.
Benötigt Argos Engine einen API-Schlüssel?
In diesem Eintrag ist keine erforderliche Umgebungsvariable aufgeführt.
Welchen Transport verwendet Argos Engine?
Argos Engine läuft lokal über stdio.
Wann wurde Argos Engine zuletzt geprüft?
Zuletzt geprüft: 2026-10-28. Token-efficient, local-first web research MCP server in Rust: keyless multi-profile metasearch. Built on FlojoMCP.