opsveritas-sdk 0.1.3 → 0.2.0

package/README.md

@@ -0,0 +1,65 @@
+# opsveritas-sdk
+
+Monitor your AI agents in **three lines of code**. Tracks tokens, cost, latency, model, and silent failures (200 OK with empty output) — and routes alerts to Slack / Email / Teams via [OpsVeritas AI Agents Control Tower](https://agents.opsveritas.com).
+
+```bash
+npm install opsveritas-sdk
+```
+
+## Quick start (3 lines)
+
+```ts
+import { OpsVeritas } from 'opsveritas-sdk';
+
+OpsVeritas.init('<your-ingest-key>');                 // key from Settings → Integrations
+const client = OpsVeritas.wrap(new OpenAI(), { agentName: 'Support Bot' });
+// use `client` exactly as before — runs appear in your dashboard automatically
+```
+
+Works the same for Anthropic and Gemini clients. Prefer manual control? Wrap a function:
+
+```ts
+await OpsVeritas.trace('Nightly Report', async () => runReport());
+```
+
+## What data is sent
+
+By design the SDK sends **metadata only, plus a short output snippet** — never your prompts/inputs:
+
+| Sent | Detail |
+|------|--------|
+| ✅ Metadata | agent name, status, timestamps, duration, token counts, model, cost, tool-call count |
+| ⚠️ Output snippet | first 300 chars of the response (`output_summary`) — powers silent-failure detection |
+| ⚠️ Error message | the exception text, if a call fails |
+| ❌ Prompts / inputs | **never sent** — only token counts |
+
+### Metadata-only mode (for regulated / client data)
+
+Drop the output snippet and redact error text so **no response content ever leaves your environment** — token/cost/latency metadata still flows:
+
+```ts
+OpsVeritas.init('<your-ingest-key>', { metadataOnly: true });
+```
+
+Or set the environment variable:
+
+```bash
+OPSVERITAS_METADATA_ONLY=true
+```
+
+## Reliability
+
+Telemetry is **non-blocking and fire-and-forget** — it never throws into your code and never slows your agent. If the ingest endpoint is briefly unreachable, sends are **retried with backoff** and **buffered in memory** (bounded), then flushed on the next event — so a transient outage doesn't lose telemetry.
+
+## Configuration
+
+```ts
+OpsVeritas.init(apiKey, {
+  endpoint,      // optional — defaults to https://agents.opsveritas.com
+  metadataOnly,  // optional — default false; when true, no response content is sent
+});
+```
+
+## License
+
+MIT

package/dist/index.d.mts

@@ -1,5 +1,6 @@
 declare function init(apiKey: string, options?: {
     endpoint?: string;
+    metadataOnly?: boolean;
 }): void;
 
 declare function run<T>(agentName: string, fn: () => Promise<T>, opts?: {

package/dist/index.d.ts

@@ -1,5 +1,6 @@
 declare function init(apiKey: string, options?: {
     endpoint?: string;
+    metadataOnly?: boolean;
 }): void;
 
 declare function run<T>(agentName: string, fn: () => Promise<T>, opts?: {

package/dist/index.js

@@ -35,7 +35,9 @@ function init(apiKey, options) {
   if (!apiKey || typeof apiKey !== "string") throw new Error("[OpsVeritas] apiKey is required");
   _config = {
     apiKey,
-    endpoint: (options?.endpoint ?? "https://agents.opsveritas.com").replace(/\/$/, "")
+    endpoint: (options?.endpoint ?? "https://agents.opsveritas.com").replace(/\/$/, ""),
+    // Opt-in via init() or the OPSVERITAS_METADATA_ONLY=true env var. Defaults off (unchanged behavior).
+    metadataOnly: options?.metadataOnly ?? (typeof process !== "undefined" && process.env?.OPSVERITAS_METADATA_ONLY === "true")
   };
 }
 function getConfig() {
@@ -47,19 +49,77 @@ function getConfig() {
 var import_async_hooks = require("async_hooks");
 
 // src/http.ts
-async function sendExecution(payload) {
-  const { apiKey, endpoint } = getConfig();
+var RETRY_BACKOFF_MS = [500, 2e3, 5e3];
+var MAX_BUFFER = 200;
+var buffer = [];
+var flushing = false;
+function sleep(ms) {
+  return new Promise((resolve) => {
+    const t = setTimeout(resolve, ms);
+    if (typeof t?.unref === "function") t.unref();
+  });
+}
+function scrubForMetadataOnly(payload) {
+  return {
+    ...payload,
+    output_summary: void 0,
+    error_message: payload.error_message != null ? "[redacted \u2014 metadata-only mode]" : payload.error_message
+  };
+}
+async function postOnce(endpoint, apiKey, body) {
   try {
-    await fetch(`${endpoint}/webhooks/agent-execution`, {
+    const res = await fetch(`${endpoint}/webhooks/agent-execution`, {
       method: "POST",
-      headers: {
-        "Content-Type": "application/json",
-        "x-opsveritas-key": apiKey
-      },
-      body: JSON.stringify(payload)
+      headers: { "Content-Type": "application/json", "x-opsveritas-key": apiKey },
+      body: JSON.stringify(body)
     });
+    if (res.ok) return "ok";
+    if (res.status >= 400 && res.status < 500) return "drop";
+    return "retry";
+  } catch {
+    return "retry";
+  }
+}
+async function deliver(endpoint, apiKey, body) {
+  for (let attempt = 0; ; attempt++) {
+    const outcome = await postOnce(endpoint, apiKey, body);
+    if (outcome === "ok" || outcome === "drop") return true;
+    if (attempt >= RETRY_BACKOFF_MS.length) return false;
+    await sleep(RETRY_BACKOFF_MS[attempt]);
+  }
+}
+function enqueue(body) {
+  buffer.push(body);
+  while (buffer.length > MAX_BUFFER) buffer.shift();
+}
+async function flush(endpoint, apiKey) {
+  if (flushing || buffer.length === 0) return;
+  flushing = true;
+  try {
+    const pending = buffer.splice(0, buffer.length);
+    for (let i = 0; i < pending.length; i++) {
+      const ok = await deliver(endpoint, apiKey, pending[i]);
+      if (!ok) {
+        for (let j = pending.length - 1; j >= i; j--) buffer.unshift(pending[j]);
+        break;
+      }
+    }
+  } finally {
+    flushing = false;
+  }
+}
+async function sendExecution(payload) {
+  let cfg;
+  try {
+    cfg = getConfig();
   } catch {
+    return;
   }
+  const { apiKey, endpoint, metadataOnly } = cfg;
+  const body = metadataOnly ? scrubForMetadataOnly(payload) : payload;
+  void flush(endpoint, apiKey);
+  const ok = await deliver(endpoint, apiKey, body);
+  if (!ok) enqueue(body);
 }
 
 // src/context.ts
@@ -111,6 +171,10 @@ async function run(agentName, fn, opts) {
 // src/pricing.ts
 var PRICING = {
   // ── OpenAI ─────────────────────────────────────────────────────────────────
+  // GPT-5.x family — most-specific-first (5.4-nano before 5.4)
+  "gpt-5.5": [5, 30],
+  "gpt-5.4-nano": [0.2, 1.25],
+  "gpt-5.4": [2.5, 15],
   // GPT-4.1 family (April 2025) — must be before 'gpt-4' to avoid wrong match
   "gpt-4.1-nano": [0.1, 0.4],
   "gpt-4.1-mini": [0.4, 1.6],
@@ -130,9 +194,9 @@ var PRICING = {
   "o1": [15, 60],
   // ── Anthropic ──────────────────────────────────────────────────────────────
   // Claude 4 family (versioned IDs matched by substring, e.g. claude-opus-4-8 → claude-opus-4)
-  "claude-opus-4": [15, 75],
+  "claude-opus-4": [5, 25],
   "claude-sonnet-4": [3, 15],
-  "claude-haiku-4": [0.8, 4],
+  "claude-haiku-4": [1, 5],
   // Claude 3.7
   "claude-3-7-sonnet": [3, 15],
   // Claude 3.5
@@ -163,26 +227,48 @@ var PRICING = {
   // ── Meta Llama 4 (Groq / Together) ────────────────────────────────────────
   "llama-4-maverick": [0.5, 0.77],
   "llama-4-scout": [0.11, 0.34],
+  // ── Moonshot (Groq) ────────────────────────────────────────────────────────
+  "kimi-k2": [1, 3],
   // ── Mistral ────────────────────────────────────────────────────────────────
+  "pixtral-large": [2, 6],
+  "ministral-3b": [0.04, 0.04],
   "mistral-large": [2, 6],
   "mistral-small": [0.1, 0.3],
+  "devstral": [0.1, 0.3],
   "codestral": [0.2, 0.6],
   "mistral-7b": [0.25, 0.25],
   "mixtral-8x22b": [0.9, 0.9],
   // ── DeepSeek ───────────────────────────────────────────────────────────────
+  "deepseek-v4-flash": [0.14, 0.28],
+  "deepseek-v4-pro": [0.435, 0.87],
   "deepseek-reasoner": [0.55, 2.19],
+  "deepseek-r1": [0.55, 2.19],
   "deepseek-chat": [0.27, 1.1],
   "deepseek-coder": [0.27, 1.1],
   // ── Cohere ─────────────────────────────────────────────────────────────────
+  "command-a": [2.5, 10],
   "command-r-plus": [2.5, 10],
   "command-r": [0.15, 0.6],
   "command": [0.15, 0.6],
   // ── Qwen (Alibaba) ─────────────────────────────────────────────────────────
+  "qwen3-235b": [1.2, 6],
+  "qwen3-32b": [0.45, 1.8],
+  "qwen3-14b": [0.2, 0.8],
+  "qwen3-8b": [0.1, 0.4],
+  "qwen3-4b": [0.05, 0.2],
   "qwen-2.5-72b": [0.9, 0.9],
   "qwen-2.5-7b": [0.1, 0.1],
+  "qwen-flash": [0.05, 0.4],
+  "qwen-max": [4, 12],
+  "qwen-plus": [0.8, 2],
+  "qwen-turbo": [0.2, 0.6],
   // ── Google Gemini ──────────────────────────────────────────────────────────
+  "gemini-3.5-flash": [1.5, 9],
+  "gemini-3.1-pro": [2, 12],
   "gemini-2.5-pro": [1.25, 10],
+  "gemini-2.5-flash-lite": [0.075, 0.3],
   "gemini-2.5-flash": [0.15, 0.6],
+  "gemini-2.0-flash-lite": [0.075, 0.3],
   "gemini-2.0-flash": [0.1, 0.4],
   "gemini-1.5-pro": [1.25, 5],
   "gemini-1.5-flash": [0.075, 0.3]

package/dist/index.mjs

@@ -4,7 +4,9 @@ function init(apiKey, options) {
   if (!apiKey || typeof apiKey !== "string") throw new Error("[OpsVeritas] apiKey is required");
   _config = {
     apiKey,
-    endpoint: (options?.endpoint ?? "https://agents.opsveritas.com").replace(/\/$/, "")
+    endpoint: (options?.endpoint ?? "https://agents.opsveritas.com").replace(/\/$/, ""),
+    // Opt-in via init() or the OPSVERITAS_METADATA_ONLY=true env var. Defaults off (unchanged behavior).
+    metadataOnly: options?.metadataOnly ?? (typeof process !== "undefined" && process.env?.OPSVERITAS_METADATA_ONLY === "true")
   };
 }
 function getConfig() {
@@ -16,19 +18,77 @@ function getConfig() {
 import { AsyncLocalStorage } from "async_hooks";
 
 // src/http.ts
-async function sendExecution(payload) {
-  const { apiKey, endpoint } = getConfig();
+var RETRY_BACKOFF_MS = [500, 2e3, 5e3];
+var MAX_BUFFER = 200;
+var buffer = [];
+var flushing = false;
+function sleep(ms) {
+  return new Promise((resolve) => {
+    const t = setTimeout(resolve, ms);
+    if (typeof t?.unref === "function") t.unref();
+  });
+}
+function scrubForMetadataOnly(payload) {
+  return {
+    ...payload,
+    output_summary: void 0,
+    error_message: payload.error_message != null ? "[redacted \u2014 metadata-only mode]" : payload.error_message
+  };
+}
+async function postOnce(endpoint, apiKey, body) {
   try {
-    await fetch(`${endpoint}/webhooks/agent-execution`, {
+    const res = await fetch(`${endpoint}/webhooks/agent-execution`, {
       method: "POST",
-      headers: {
-        "Content-Type": "application/json",
-        "x-opsveritas-key": apiKey
-      },
-      body: JSON.stringify(payload)
+      headers: { "Content-Type": "application/json", "x-opsveritas-key": apiKey },
+      body: JSON.stringify(body)
     });
+    if (res.ok) return "ok";
+    if (res.status >= 400 && res.status < 500) return "drop";
+    return "retry";
+  } catch {
+    return "retry";
+  }
+}
+async function deliver(endpoint, apiKey, body) {
+  for (let attempt = 0; ; attempt++) {
+    const outcome = await postOnce(endpoint, apiKey, body);
+    if (outcome === "ok" || outcome === "drop") return true;
+    if (attempt >= RETRY_BACKOFF_MS.length) return false;
+    await sleep(RETRY_BACKOFF_MS[attempt]);
+  }
+}
+function enqueue(body) {
+  buffer.push(body);
+  while (buffer.length > MAX_BUFFER) buffer.shift();
+}
+async function flush(endpoint, apiKey) {
+  if (flushing || buffer.length === 0) return;
+  flushing = true;
+  try {
+    const pending = buffer.splice(0, buffer.length);
+    for (let i = 0; i < pending.length; i++) {
+      const ok = await deliver(endpoint, apiKey, pending[i]);
+      if (!ok) {
+        for (let j = pending.length - 1; j >= i; j--) buffer.unshift(pending[j]);
+        break;
+      }
+    }
+  } finally {
+    flushing = false;
+  }
+}
+async function sendExecution(payload) {
+  let cfg;
+  try {
+    cfg = getConfig();
   } catch {
+    return;
   }
+  const { apiKey, endpoint, metadataOnly } = cfg;
+  const body = metadataOnly ? scrubForMetadataOnly(payload) : payload;
+  void flush(endpoint, apiKey);
+  const ok = await deliver(endpoint, apiKey, body);
+  if (!ok) enqueue(body);
 }
 
 // src/context.ts
@@ -80,6 +140,10 @@ async function run(agentName, fn, opts) {
 // src/pricing.ts
 var PRICING = {
   // ── OpenAI ─────────────────────────────────────────────────────────────────
+  // GPT-5.x family — most-specific-first (5.4-nano before 5.4)
+  "gpt-5.5": [5, 30],
+  "gpt-5.4-nano": [0.2, 1.25],
+  "gpt-5.4": [2.5, 15],
   // GPT-4.1 family (April 2025) — must be before 'gpt-4' to avoid wrong match
   "gpt-4.1-nano": [0.1, 0.4],
   "gpt-4.1-mini": [0.4, 1.6],
@@ -99,9 +163,9 @@ var PRICING = {
   "o1": [15, 60],
   // ── Anthropic ──────────────────────────────────────────────────────────────
   // Claude 4 family (versioned IDs matched by substring, e.g. claude-opus-4-8 → claude-opus-4)
-  "claude-opus-4": [15, 75],
+  "claude-opus-4": [5, 25],
   "claude-sonnet-4": [3, 15],
-  "claude-haiku-4": [0.8, 4],
+  "claude-haiku-4": [1, 5],
   // Claude 3.7
   "claude-3-7-sonnet": [3, 15],
   // Claude 3.5
@@ -132,26 +196,48 @@ var PRICING = {
   // ── Meta Llama 4 (Groq / Together) ────────────────────────────────────────
   "llama-4-maverick": [0.5, 0.77],
   "llama-4-scout": [0.11, 0.34],
+  // ── Moonshot (Groq) ────────────────────────────────────────────────────────
+  "kimi-k2": [1, 3],
   // ── Mistral ────────────────────────────────────────────────────────────────
+  "pixtral-large": [2, 6],
+  "ministral-3b": [0.04, 0.04],
   "mistral-large": [2, 6],
   "mistral-small": [0.1, 0.3],
+  "devstral": [0.1, 0.3],
   "codestral": [0.2, 0.6],
   "mistral-7b": [0.25, 0.25],
   "mixtral-8x22b": [0.9, 0.9],
   // ── DeepSeek ───────────────────────────────────────────────────────────────
+  "deepseek-v4-flash": [0.14, 0.28],
+  "deepseek-v4-pro": [0.435, 0.87],
   "deepseek-reasoner": [0.55, 2.19],
+  "deepseek-r1": [0.55, 2.19],
   "deepseek-chat": [0.27, 1.1],
   "deepseek-coder": [0.27, 1.1],
   // ── Cohere ─────────────────────────────────────────────────────────────────
+  "command-a": [2.5, 10],
   "command-r-plus": [2.5, 10],
   "command-r": [0.15, 0.6],
   "command": [0.15, 0.6],
   // ── Qwen (Alibaba) ─────────────────────────────────────────────────────────
+  "qwen3-235b": [1.2, 6],
+  "qwen3-32b": [0.45, 1.8],
+  "qwen3-14b": [0.2, 0.8],
+  "qwen3-8b": [0.1, 0.4],
+  "qwen3-4b": [0.05, 0.2],
   "qwen-2.5-72b": [0.9, 0.9],
   "qwen-2.5-7b": [0.1, 0.1],
+  "qwen-flash": [0.05, 0.4],
+  "qwen-max": [4, 12],
+  "qwen-plus": [0.8, 2],
+  "qwen-turbo": [0.2, 0.6],
   // ── Google Gemini ──────────────────────────────────────────────────────────
+  "gemini-3.5-flash": [1.5, 9],
+  "gemini-3.1-pro": [2, 12],
   "gemini-2.5-pro": [1.25, 10],
+  "gemini-2.5-flash-lite": [0.075, 0.3],
   "gemini-2.5-flash": [0.15, 0.6],
+  "gemini-2.0-flash-lite": [0.075, 0.3],
   "gemini-2.0-flash": [0.1, 0.4],
   "gemini-1.5-pro": [1.25, 5],
   "gemini-1.5-flash": [0.075, 0.3]

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "opsveritas-sdk",
-  "version": "0.1.3",
-  "description": "Monitor your AI agents with 2 lines of code",
+  "version": "0.2.0",
+  "description": "Monitor your AI agents in 3 lines of code — tokens, cost, latency, and silent failures",
   "main": "dist/index.js",
   "module": "dist/index.mjs",
   "types": "dist/index.d.ts",
@@ -12,7 +12,9 @@
       "types": "./dist/index.d.ts"
     }
   },
-  "files": ["dist"],
+  "files": [
+    "dist"
+  ],
   "scripts": {
     "build": "tsup src/index.ts --format esm,cjs --dts --clean",
     "dev": "tsup src/index.ts --format esm,cjs --dts --watch",
@@ -31,9 +33,21 @@
     "openai": ">=4.0.0"
   },
   "peerDependenciesMeta": {
-    "@anthropic-ai/sdk": { "optional": true },
-    "openai": { "optional": true }
+    "@anthropic-ai/sdk": {
+      "optional": true
+    },
+    "openai": {
+      "optional": true
+    }
   },
-  "keywords": ["ai", "agents", "monitoring", "observability", "openai", "anthropic", "langchain"],
+  "keywords": [
+    "ai",
+    "agents",
+    "monitoring",
+    "observability",
+    "openai",
+    "anthropic",
+    "langchain"
+  ],
   "license": "MIT"
 }