coinopai-mcp 1.2.7 → 1.2.9

package/README.md

@@ -7,21 +7,21 @@
 [![payments](https://img.shields.io/badge/payments-x402%20USDC-0052FF)](https://x402.org)
 [![network](https://img.shields.io/badge/network-Base-0052FF)](https://base.org)
 
-**Directional market intelligence with built-in outcome verification.**
+**Auditable market context with calibrated forecast ranges and outcome verification.**
 
 **Source:** https://github.com/forgemeshlabs/coinopai-mcp
 
-An MCP server that lets AI agents buy Kronos market intelligence with [x402](https://x402.org) micropayments on Base. Kronos does not just generate signals: every decision gets a `decision_id`, and every `decision_id` can be audited against later market prices.
+An MCP server that lets AI agents buy Kronos market context with [x402](https://x402.org) micropayments on Base. Kronos is not a buy/sell oracle: it gives agents calibrated ranges, risk context, decision journals, and audit records they can verify.
 
 > This repo is the MCP client layer; paid intelligence is served from hosted CoinOpAI x402 endpoints.
 
-> Wrong predictions are shown too. That's the point.
+> Weak calls and wrong calls are shown too. That's the point.
 
 ## Why Kronos Is Different
 
-Most signal APIs stop after making a prediction. Kronos assigns a `decision_id` and lets every decision be audited against future market behavior.
+Most market APIs stop after returning a direction. Kronos leads with calibrated ranges and risk context, then assigns a `decision_id` so every decision can be audited against future market behavior.
 
-The moat is the verification loop:
+The moat is the audit loop:
 
 ```
 preflight -> decision -> audit
@@ -61,45 +61,46 @@ The agent calls a tool → the MCP server receives an `HTTP 402` → automatical
 
 ---
 
-## The Verified Loop ($0.27/cycle)
+## The Auditable Loop ($0.27/cycle)
 
 ```
 check_trade_preflight  ──→  get_crypto_decision  ──→  [wait 1h]  ──→  audit_trade_decision
        $0.05                      $0.15                                      $0.07
 
-   Is now allowed?           CONSIDER_LONG/SHORT              GOOD_DECISION
-   Cooldown check?           NO_ACTION                        BAD_DIRECTION
-   Regime ok?                + decision_id                    NOISE
-   Signal strength?          + next_step hint                 + pnl_pct
+   Market state?             Directional context              Outcome record
+   Cooldown context?         Confidence context               Direction held?
+   Regime context?           + decision_id                    Verdict
+   Data freshness?           + audit hint                     + pnl_pct
 ```
 
-Every decision is self-verifying. The `decision_id` links the prediction to the outcome. The audit fetches real market prices and produces a verdict. Nothing is hidden.
+Every decision is self-verifying. The `decision_id` links the setup to the outcome. The audit fetches real market prices and produces a verdict. Nothing is hidden.
 
 ## Current Status
 
 **Live**
-- Signals
+- Model context
 - Risk assessment
-- Decision guidance
+- Decision journaling
 - Outcome verification
+- Forecast: conformally-calibrated 80% price range (~0.80 empirical coverage)
 
 **Research**
-- Forecast intelligence: validation in progress
+- Directional edge: none demonstrated in backtest (~51% accurate) — the calibrated range is the validated product, not the direction
 - Forecast vs execution agreement analysis: collecting evidence
 
-Forecast outputs are not exposed publicly. Validation comes first.
+The calibrated forecast range is live and validated (conformal, ~0.80 coverage). Directional values are supporting context, not standalone trade instructions. Use them with the calibrated range, risk state, and audit record.
 
-## Signal Scale
+## Directional Context
 
-| Signal | Meaning |
+| Value | Meaning |
 |--------|---------|
-| Positive | Bullish directional read |
-| Negative | Bearish directional read |
+| Positive | Bullish model context |
+| Negative | Bearish model context |
 | 0.00-0.01 | Weak magnitude |
 | 0.01-0.03 | Moderate magnitude |
 | 0.03+ | Strong magnitude |
 
-Signals are probabilistic model outputs, not guarantees or human recommendations.
+Directional values are supporting context, not guarantees, human recommendations, or standalone trade instructions. Use them with the calibrated range, risk state, and audit record.
 
 ---
 
@@ -111,7 +112,7 @@ Signals are probabilistic model outputs, not guarantees or human recommendations
   "allowed": true,
   "symbol": "BTC/USD",
   "market_state": "NORMAL",
-  "signal_strength": 0.72,
+  "signal_strength": "weak_or_mixed",
   "regime": "TREND",
   "cooldown_remaining_seconds": 0
 }
@@ -121,10 +122,15 @@ Signals are probabilistic model outputs, not guarantees or human recommendations
 ```json
 {
   "symbol": "BTC/USD",
-  "suggested_action": "CONSIDER_LONG",
-  "confidence": 0.72,
+  "directional_bias": "upward",
+  "confidence": 0.514,
+  "compliance_mode": "market_intelligence_only",
   "regime": "TREND",
   "decision_id": "a3f8c1d2-9472-4dfe-b459-5df17b282614",
+  "directional_edge": "none_demonstrated",
+  "why_not_high": [
+    "Directional confidence is capped by observed historical accuracy, not boosted by signal magnitude."
+  ],
   "next_step": "Call audit_trade_decision with this decision_id after 1h using window=1h"
 }
 ```
@@ -139,15 +145,7 @@ Signals are probabilistic model outputs, not guarantees or human recommendations
 }
 ```
 
-**Live results from real runs:**
-
-| Symbol | Decision | Confidence | 1h Outcome | Verdict |
-|--------|----------|-----------|------------|---------|
-| XRP | SHORT | 1.0 | -0.54% | ✅ GOOD_DECISION |
-| ETH | LONG | 0.68 | -0.32% | ❌ BAD_DIRECTION |
-| BTC | LONG | 0.40 | +0.01% | — NOISE |
-
-It gets some right. It gets some wrong. The loop makes both visible.
+Audit verdicts include `GOOD_DECISION`, `BAD_DIRECTION`, `NOISE`, `NO_ACTION_TAKEN`, and `PENDING`. Kronos gets some right, gets some wrong, and exposes both through the same record.
 
 ---
 
@@ -155,11 +153,13 @@ It gets some right. It gets some wrong. The loop makes both visible.
 
 | Tool | What it does | Cost | Affiliate |
 |------|-------------|------|-----------|
-| `check_trade_preflight` | Gate check: market allowed, cooldown, regime, signal strength | $0.05 | ✓ |
-| `get_crypto_decision` | CONSIDER\_LONG/SHORT/NO\_ACTION + `decision_id` | $0.15 | ✓ |
+| `check_trade_preflight` | Gate check: market allowed, cooldown, regime, model context | $0.05 | ✓ |
+| `get_crypto_decision` | Probabilistic decision journal + `decision_id` | $0.15 | ✓ |
 | `audit_trade_decision` | Verify against real prices: verdict + PnL% | $0.07 | ✓ |
-| `get_crypto_signals` | Directional market intelligence for BTC, ETH, SOL, XRP, ADA | $0.05 | ✓ |
-| `get_crypto_signal_history` | Up to 168h of signal history for analysis | $0.05 | ✓ |
+| `get_crypto_signals` | Model context for BTC, ETH, SOL, XRP, ADA | $0.05 | ✓ |
+| `get_crypto_signal_history` | Up to 168h of context history for analysis | $0.05 | ✓ |
+| `get_crypto_forecast` | Conformally-calibrated 80% price range (~0.80 empirical coverage) for BTC, ETH, SOL, XRP, ADA | $0.05 | ✓ |
+| `review_signal_anomaly` | Score signal features for unusual conditions; returns review labels, drivers, and component scores | $0.07 | — |
 | `get_crypto_risk` | Market risk state and cooldown context | $0.02 | — |
 | `search_agent_automations` | Search 819 agent automation prompts | $0.01 | — |
 | `get_agent_automation` | Full prompt + workflow steps by slug | $0.01 | — |
@@ -189,7 +189,7 @@ Add to `~/.claude/settings.json`:
 }
 ```
 
-Restart Claude Code. The 9 tools appear automatically.
+Restart Claude Code. The tools appear automatically.
 
 ### Claude Desktop
 
@@ -237,16 +237,27 @@ Prepared MCP Registry identity: `io.github.forgemeshlabs/coinopai-mcp`. Refresh
 const pre = await mcp.call("check_trade_preflight", { symbol: "BTC" })
 if (!pre.allowed) return  // cooldown, bad regime, or stale data
 
-// Step 2 — get decision ($0.15)
+// Step 2 — get decision journal ($0.15)
 const dec = await mcp.call("get_crypto_decision", { symbol: "BTC" })
-if (dec.suggested_action === "NO_ACTION") return
 
 // Store the decision_id — you'll need it to close the loop
-const { decision_id, suggested_action, confidence } = dec
+const { decision_id, directional_bias, confidence } = dec
 
-// Step 3 — act on the decision here...
+// Optional — review a feature set for anomaly context ($0.07)
+const anomaly = await mcp.call("review_signal_anomaly", {
+  symbol: "BTC",
+  window: "24h",
+  features: {
+    price_change: 0.018,
+    volume_change: 0.42,
+    volatility: 0.031,
+    signal_confidence: 72,
+    risk_score: 31
+  }
+})
+// review_label: "normal_review" | "review" | "elevated_review" | "critical_review"
 
-// Step 4 — audit 1 hour later ($0.07)
+// Step 3 — audit 1 hour later ($0.07)
 const audit = await mcp.call("audit_trade_decision", {
   decision_id,
   window: "1h"
@@ -340,7 +351,7 @@ Normal x402 flow — CoinOpAI receives 100% of the listed price. Nothing changes
 
 ## Disclaimer
 
-Decision outputs are probabilistic signals for experimental automated workflows only. Not financial advice. Early system with a small sample size. Results will vary. Never risk capital you can't afford to lose.
+Decision outputs are probabilistic context and journal entries for experimental automated workflows only. Not financial advice. Directional bias alone has not been validated as a standalone trading strategy. Results will vary. Never risk capital you can't afford to lose.
 
 ---
 

package/index.js

@@ -68,7 +68,7 @@ const TOOLS = [
   },
   {
     name: "get_crypto_signals",
-    description: "Latest Kronos directional market intelligence for BTC, ETH, SOL, XRP, ADA. Positive = bullish, negative = bearish; magnitude under 0.01 is weak, 0.01-0.03 moderate, 0.03+ strong. Costs $0.05 USDC.",
+    description: "Latest Kronos model context for BTC, ETH, SOL, XRP, ADA. Directional values are supporting context, not standalone trade instructions; use them with the calibrated range, risk state, and audit record. Costs $0.05 USDC.",
     inputSchema: {
       type: "object",
       properties: {
@@ -83,7 +83,7 @@ const TOOLS = [
   },
   {
     name: "get_crypto_signal_history",
-    description: "Recent Kronos signal history for BTC/ETH/SOL/XRP/ADA. Use it to inspect directional context before or after a decision. Costs $0.05 USDC.",
+    description: "Recent Kronos context history for BTC/ETH/SOL/XRP/ADA. Use it to inspect model context and freshness before or after a decision. Costs $0.05 USDC.",
     inputSchema: {
       type: "object",
       properties: {
@@ -94,7 +94,7 @@ const TOOLS = [
   },
   {
     name: "get_crypto_decision",
-    description: "Get probabilistic decision guidance from Kronos, then verify it. Returns CONSIDER_LONG/SHORT/NO_ACTION with confidence, regime, and a decision_id. Call audit_trade_decision with that ID after the evaluation window to see whether the direction held. Costs $0.15 USDC.",
+    description: "Create a market-intelligence journal entry from Kronos context. Returns directional_bias, confidence context, compliance metadata, regime, anomaly/calibration context, and a decision_id. Call audit_trade_decision with that ID after the evaluation window to see what happened. Costs $0.15 USDC.",
     inputSchema: {
       type: "object",
       properties: {
@@ -106,7 +106,7 @@ const TOOLS = [
   },
   {
     name: "check_trade_preflight",
-    description: "Step 1 of the verified decision loop. Checks market state, cooldown, data freshness, and signal strength before calling get_crypto_decision. Costs $0.05 USDC.",
+    description: "Step 1 of the auditable decision loop. Checks market state, cooldown, data freshness, and model context before calling get_crypto_decision. Costs $0.05 USDC.",
     inputSchema: {
       type: "object",
       properties: {
@@ -131,7 +131,7 @@ const TOOLS = [
   },
   {
     name: "get_crypto_forecast",
-    description: "Conformally-calibrated price forecast: an honest 80% prediction interval (range_80, ~0.80 empirical coverage) plus point return and upside probability for BTC/ETH/SOL/XRP/ADA. The directional point has no demonstrated backtest edge (~51% accurate) — the calibrated range is the validated product. Costs $0.05 USDC.",
+    description: "Conformally-calibrated price forecast: an honest 80% prediction interval (range_80, ~0.80 empirical coverage) plus point return and upside probability for BTC/ETH/SOL/XRP/ADA. Directional bias is supporting context; the calibrated range is the validated product. Costs $0.05 USDC.",
     inputSchema: {
       type: "object",
       properties: {
@@ -139,6 +139,22 @@ const TOOLS = [
         affiliate_id: { type: "string", description: "Optional Pyrimid affiliate ID (af_xxxxx). Affiliate earns a commission from within the listed price — no extra cost to you." }
       }
     }
+  },
+  {
+    name: "review_signal_anomaly",
+    description: "Score a market signal feature set for unusual conditions before downstream analysis. Returns anomaly_score, anomaly_level, review_label, drivers, component scores, and market-intelligence disclaimers. Not financial advice and not a market activity instruction. Costs $0.07 USDC.",
+    inputSchema: {
+      type: "object",
+      properties: {
+        symbol: { type: "string", description: "Symbol to review, e.g. BTC, ETH, SOL, XRP, ADA, AAPL, SPY" },
+        window: { type: "string", description: "Observation window label, e.g. 24h (default: 24h)" },
+        features: {
+          type: "object",
+          description: "Numeric feature values to score, such as price_change, volume_change, volatility, signal_confidence, risk_score, social_velocity, or onchain_velocity."
+        }
+      },
+      required: ["symbol", "features"]
+    }
   }
 ];
 
@@ -192,10 +208,14 @@ async function callPaid(ctx, path, affiliateId, opts = {}) {
   const baseUrl = opts.baseUrl || BASE_URL;
   const pyrimidProducts = opts.pyrimidProducts || PYRIMID_PRODUCTS;
   const pathname = new URL(path, baseUrl).pathname;
+  const method = opts.method || "GET";
+  const body = opts.body;
+  const bodyHeaders = body ? { "Content-Type": "application/json" } : {};
+  const bodyInit = body ? { body: JSON.stringify(body) } : {};
 
   // Use Pyrimid affiliate flow when affiliate_id present and product is registered.
   // Fall back to direct x402 if the affiliate route is unavailable or not yet cataloged.
-  if (affiliateId && pyrimidProducts[pathname]) {
+  if (method === "GET" && affiliateId && pyrimidProducts[pathname]) {
     try {
       return await callPyrimid(account, path, affiliateId, baseUrl, pyrimidProducts);
     } catch (_) {
@@ -206,7 +226,7 @@ async function callPaid(ctx, path, affiliateId, opts = {}) {
   // Standard x402 EIP-3009 flow
   const url = baseUrl + path;
   const extraHeaders = affiliateId ? { "X-Affiliate-ID": affiliateId } : {};
-  const res = await fetch(url, { headers: extraHeaders });
+  const res = await fetch(url, { method, headers: { ...bodyHeaders, ...extraHeaders }, ...bodyInit });
 
   if (res.status === 402) {
     let body;
@@ -216,7 +236,9 @@ async function callPaid(ctx, path, affiliateId, opts = {}) {
     );
     const paymentPayload = await httpClient.createPaymentPayload(paymentRequired);
     const paidRes = await fetch(url, {
-      headers: { ...httpClient.encodePaymentSignatureHeader(paymentPayload), ...extraHeaders },
+      method,
+      headers: { ...bodyHeaders, ...httpClient.encodePaymentSignatureHeader(paymentPayload), ...extraHeaders },
+      ...bodyInit,
     });
     if (!paidRes.ok) {
       const errBody = await paidRes.text().catch(() => paidRes.statusText);
@@ -240,7 +262,7 @@ async function main() {
   }
 
   const server = new Server(
-    { name: "coinopai-mcp", version: "1.2.7" },
+    { name: "coinopai-mcp", version: "1.2.9" },
     { capabilities: { tools: {} } }
   );
 
@@ -286,6 +308,16 @@ async function main() {
         case "get_crypto_forecast":
           data = await callPaid(paymentContext, `/api/kronos/forecast?symbol=${encodeURIComponent(args.symbol || "BTC")}`, affiliateId);
           break;
+        case "review_signal_anomaly":
+          data = await callPaid(paymentContext, "/api/anomaly", null, {
+            method: "POST",
+            body: {
+              symbol: args.symbol || "BTC",
+              window: args.window || "24h",
+              features: args.features || {},
+            },
+          });
+          break;
         default:
           throw new Error("Unknown tool: " + name);
       }

package/package.json

@@ -1,8 +1,8 @@
 {
   "name": "coinopai-mcp",
-  "version": "1.2.7",
+  "version": "1.2.9",
   "mcpName": "io.github.forgemeshlabs/coinopai-mcp",
-  "description": "Directional crypto market intelligence MCP with x402 payments, decision guidance, and built-in outcome verification",
+  "description": "Auditable market intelligence MCP with x402 payments, calibrated forecast ranges, anomaly review, risk context, and outcome audits",
   "main": "index.js",
   "bin": {
     "coinopai-mcp": "index.js"
@@ -28,6 +28,9 @@
     "forgemesh",
     "crypto",
     "signals",
+    "forecast",
+    "risk",
+    "audit",
     "coinbase",
     "base",
     "usdc",

package/server.json

@@ -1,17 +1,17 @@
 {
   "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
   "name": "io.github.forgemeshlabs/coinopai-mcp",
-  "description": "Directional crypto market intelligence with decision guidance, outcome verification, and x402 micropayments on Base.",
+  "description": "Auditable market intelligence with calibrated forecast ranges, anomaly review, outcome verification, and x402 micropayments on Base.",
   "repository": {
     "url": "https://github.com/forgemeshlabs/coinopai-mcp",
     "source": "github"
   },
-  "version": "1.2.6",
+  "version": "1.2.9",
   "packages": [
     {
       "registryType": "npm",
       "identifier": "coinopai-mcp",
-      "version": "1.2.6",
+      "version": "1.2.9",
       "transport": {
         "type": "stdio"
       },