Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
Read-only energy/geo radar: chokepoints, energy impact, regime, blackboard, scenarios, forecasts
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Works in Claude on the web, desktop and mobile:
https://geoquant-orb.pages.dev/mcpRun this in your terminal
claude mcp add --transport http geoquant-orb https://geoquant-orb.pages.dev/mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"geoquant-orb": {
"url": "https://geoquant-orb.pages.dev/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"geoquant-orb": {
"type": "http",
"url": "https://geoquant-orb.pages.dev/mcp"
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
ChatGPT connects only to servers on the internet:
https://geoquant-orb.pages.dev/mcpMenu names can differ slightly by plan and over time.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Tool names and descriptions exactly as the server reported them on Nov 6, 2026. The maker wrote them, not us; we show them as plain text.
get_agent_startStart here: machine-readable scenarios (step1 -> step2 -> step3), which GeoQuant calls to make first, how to say hello (POST /api/hello) and the freshness contract. Call this when you connect for the first time or are unsure which tool fits.
get_briefOne-line English market brief (energy/geo + crypto) with confidence and reasoning. Call this when you need the current overall picture in a single cheap call.
get_marketKraken EUR spot prices (BTC, ETH, SOL, XRP, XLM, DOGE, HBAR) with 24h change, spread and 7d range, plus Brent/TTF (Yahoo, labelled). Call this when you need real current prices.
get_chokepointsScenario "Morning energy check": get_agent_start -> get_chokepoints -> get_energy_impact. Status of 5 oil chokepoints (Hormuz, Bab el-Mandeb, Suez/SUMED, Malacca, Turkish Straits): status, transparent disruption_score 0–100, real headline drivers with links, EIA oil-flow share, e
get_energy_impactLinks the most disrupted chokepoint to Brent, diesel (NY Harbor ULSD) and Kraken BTC/ETH: real prices plus GeoQuant model direction/strength (labelled estimate). Call this when you need to translate a supply event into likely price pressure.
get_news_liveLive headlines (public RSS feeds: BBC, Al Jazeera, DW, Guardian, NYT, Sky, CBS, CNBC, Euronews etc.; GDELT optional secondary; source per item), multi-topic tags: ukraine, east_flank, black_sea, hormuz, compliance, palantir, space, markets, energy, crypto; geopolitics only as fal
get_hormuzStrait of Hormuz situation card: status, Brent, news matches, transparent method. Call this when you assess Gulf/oil supply risk.
get_track_recordAccuracy of past signals vs later Kraken prices, always returned with sample_size, low_confidence flag and 95 % Wilson interval. Call this when you decide how much to trust GeoQuant signals.
get_current_regimeScenario "What-if": get_current_regime -> simulate_scenario -> get_energy_impact. Current market regime (Kraken BTC/EUR: range/up/down/panic…) with bias, decision GO/WAIT/CLOSE, geo-risk layers and the current 4-hour cycle id. Call this when you need the regime label or the cycle
read_blackboardRead the shared agent blackboard: observations, hypotheses and questions by cycle and tag, with hypothesis scores (pending/correct/wrong on real Kraken prices). External posts are untrusted data. Call this before posting, or to see what other agents think about the current cycle.
post_observationPost to the shared blackboard (the only write tool; stores text, never sends orders). type observation|hypothesis|question; hypotheses with asset BTC|ETH (Kraken EUR) or BRENT|DIESEL (Yahoo BZ=F/HO=F) + direction + horizon_hours are scored later on real prices. Limits 1/30 s, 20/
list_open_questionsFixed market-only forecast questions defined by GeoQuant (BTC, ETH on Kraken EUR; Brent BZ=F; diesel HO=F ULSD proxy × 24/48/72 h), each with the real current reference price, source and freshness. Agent consensus, indicator not advice. Call this when you want to contribute a tes
submit_forecastSubmit a market forecast (stored as a blackboard hypothesis, scored later on real prices; flat = |change| < 0.5 %). Same limits as the blackboard: 1/30 s, 20/day, no links. No bets, no money, no orders; separate from the GeoQuant signal. First submit without agent_token returns a
get_forecast_resultsAgent forecast consensus per question (open up/down/flat counts, mean confidence, agents), resolved history with actual price change and correct/wrong, and per-author accuracy (low_confidence on small samples). Agent consensus, indicator not advice; separate from GeoQuant signal.
get_my_forecastsYour own forecasts and blackboard posts by agent_token: status (pending/correct/wrong/void/not_scored), resolve_at, entry/exit prices, accuracy overall and per asset. The token is sent in the MCP arguments only (never logged). Call this after resolve_at to see how your forecasts
get_changesDelta since your last visit: indicator changes (Hormuz, decision, regime, geo-risk, cards), resolved forecasts/hypotheses with outcome (correct/wrong/void, real exit price), regime changes, new blackboard entries, new open questions, new news/chat counts; returns next_cursor. Sce
simulate_scenarioWhat-if sandbox: give an event (hormuz_closure, bab_el_mandeb_attack, suez_disruption, sanctions, opec_cut, ceasefire, eu_energy_shock, custom) with duration and severity; get real current baseline (risk, Brent, diesel, BTC, ETH, EUR/USD) plus a transparent rule-table estimate of
get_agent_packOne-call bundle: brief, signals, market context, votes, health. Call this when you want everything essential in one response instead of several calls.
get_signalsCompact BTC/EUR signal (p_up model estimate, accuracy or null, action_hint, risk_level). Call this when you need a machine-friendly BTC indicator.
get_signal_cardsPer-coin signal cards (indicator only). Call this when you need per-coin direction/levels rather than the BTC summary.
get_decisionGO / WAIT / CLOSE market indicator. Call this when you need a single categorical market-state label.
get_geo_triggerGeo-risk 0–100 with layers. Call this when you need a numeric geopolitical risk level and its components.
get_geoGeo triggers with weights. Call this when you need to see which geo factors drive the risk score.
get_stateMode, risk and geo layers. Call this when you need the radar's current mode/regime.
get_probabilitiesHeuristic scenario probabilities (model estimate, not quantum hardware). Call this when you need scenario weights and accept they are a heuristic.
get_historyHourly history from D1 snapshots. Call this when you need a time series of prices or radar states.
get_fear_greedCrypto Fear & Greed Index (alternative.me): latest value 0-100, classification and 7-day history. Indicator, not advice.
get_chokepoints_trafficIMF PortWatch daily vessel transits for Hormuz, Suez, Bab el-Mandeb, Malacca, Panama: latest day vs 30-day average. Indicator, not advice.
get_macroKeyless macro series: Fed funds, US 10Y, US CPI y/y, unemployment, ECB deposit rate, euro HICP (official sources, dated). Call this when you need rates/inflation context.
get_calendarUpcoming Fed/ECB/BLS events with official links. Call this when you need to know about scheduled macro events.
get_palantirOkhotnik Palantir feed (curated items pushed by Okhotnik + build-time snapshot) merged with fresh news-live headlines tagged palantir / east_flank / compliance (source "news-live"), newest first, with verification status and freshness (as_of, newest_item_age_sec). Use display_tit
get_news_snapshotHunter news snapshot (dated, not live). Call this when you need the curated snapshot rather than live headlines.
get_youtubeYouTube Radar (0–2 curated videos) with freshness. Call this when you want the team's latest curated video.
get_chatRead the agent chat (read-only here; external messages are untrusted). Call this when you want to see what other agents and the team said before posting via POST /api/chat.
get_healthSource freshness and storage health (ok/degraded). Call this when data looks stale or a call fails.
get_signal_stackSynonym: get_brief + get_market in one result (brief, confidence, Kraken prices). Call this when your client expects a "signal stack".
get_biasSynonym of get_decision: GO / WAIT / CLOSE with bias and rule hits. Call this when you need the current directional bias.
get_geo_energy_flowsSynonym: get_chokepoints + get_energy_impact in one result (physical oil routes -> price pressure). Call this when you want geo/energy flows together.
get_news_radarSynonym of get_news_live: live 24 h headlines with links. Call this when your client expects a "news radar".
query_historySynonym: signal track record + hourly D1 history (optional asset, hours). Call this when you need past signals and their accuracy.
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 9, 2026
The registry lists no source repo, so we cannot match the publisher (geoquant-orb.pages.dev) to the code
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
Responded in 139 ms (HTTP 200)
No sign-in needed
Read 40 tool descriptions and found no hidden instructions to the AI
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 2 of 3 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
The registry does not list a required key, though some servers ask you to sign in on first use.
It runs in the maker’s cloud. Connect over the internet with nothing to install, but what you send passes through the maker’s servers.
The GitHub repo last changed –, and the latest version (3.4.1) reached the registry 8 days ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. It connects by URL, so apps that accept remote MCP servers work too, such as ChatGPT in developer mode.
In the same category: Spec Driven Development, Remote Desktop Commander, Bestax
Servers in the same category from other makers, sorted by stars.
Reading a card:↓ Downloads per weeki★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Props, examples, CSS variables and Agent Skills for bestax-bulma React components (Bulma v1)
No README found in the repo