Veqen: AI cost monitoring and protection changelog

https://veqen-app.com/llms.txt

− > Veqen is cost monitoring and protection for AI. For work and coding agents, it shows each person's and team's cost across ChatGPT, Claude, Microsoft 365 Copilot, Claude Code, Cursor, GitHub Copilot, Codex, Devin and the other agents, priced at the company's contracts, with limits pushed into each tool, misuse alerts, idle seats and chargeback. For product inference, it reads each provider's billing and per-key usage to catch stolen keys and unexplained spend within the hour, shut the key off, − - Price book as JSON: https://veqen-app.com/prices.json (version 2026-10-28)− - [Coding-agent costs explained](https://veqen-app.com/guides/coding-agent-costs.md): How Claude Code, Cursor, Copilot, Codex, Devin and the other coding tools bill, why their dashboards disagree with your invoice, and how to get one number per engineer and team.− - [What to do about a stolen AI API key](https://veqen-app.com/guides/stolen-ai-key.md): How to tell a stolen OpenAI, Anthropic or other AI provider key from a busy day, shut it off, gather the evidence a provider asks for with a refund request, and keep it from happening again.− - [How to measure LLM cost per customer](https://veqen-app.com/guides/cost-per-customer.md): Attribute every OpenAI, Anthropic and Gemini call to the customer that caused it, in Python, Node.js, the Vercel AI SDK and LangChain.− - [Measure AI agent cost per run](https://veqen-app.com/guides/ai-agent-cost-per-run.md): Add up every model call in an agent task, compare median and expensive runs, and find retry loops before choosing a per-task price.+ > Veqen is cost monitoring and protection for AI. For work and coding agents, it shows each person's and team's cost across ChatGPT, Claude, Microsoft 365 Copilot, Claude Code, Cursor, GitHub Copilot, Codex, Devin and the other agents, priced at the company's contracts, with limits pushed into each tool, misuse alerts, idle seats and chargeback. For product inference, it reads each provider's billing and per-key usage to catch stolen keys and unexplained spend within the hour, shut the key off, + - Agent spend (beta), fraud protection for AI agents that spend on cards and accounts: https://veqen-app.com/agent-spend; docs: https://veqen-app.com/docs/agent-spend.md+ - Vendors: each AI vendor's effective rate against list price and what you agreed, agreements with notice dates and commitment pacing, and savings proven before and after a renegotiation: https://veqen-app.com/procurement; docs: https://veqen-app.com/docs/rates.md, https://veqen-app.com/docs/agreements.md, https://veqen-app.com/docs/proven-savings.md+ - Price book as JSON: https://veqen-app.com/prices.json (version 2026-10-30)+ - [Agent spend](https://veqen-app.com/docs/agent-spend.md): Give each AI agent a mandate, have it ask Veqen before it spends, and match every card authorization and payment on Stripe Issuing, Ramp, Brex and Mercury to what the agent said it was doing, with approvals, incidents, card freeze, evidence and a monthly controls report.+ - [Your rate with each AI vendor](https://veqen-app.com/docs/rates.md): What each AI provider really costs as a share of its list price, whether the price you agreed is on the bill, and what each seat tool costs per person who used a seat.+ - [Agreements, renewals and commitments](https://veqen-app.com/docs/agreements.md): Record AI vendor agreements with their notice dates, commitments, volume tiers and overage, see how spending is pacing, and get reminded before each notice date.+ - [Proven savings](https://veqen-app.com/docs/proven-savings.md): Prove a renegotiated AI contract reached the bill: freeze a baseline before the new terms, then see each month what the new rate saved, apart from usage changes and the vendor's own list price changes.+ - [AI API key security: detect leaked-key spend and respond](https://veqen-app.com/guides/stolen-ai-key.md): Investigate leaked AI API keys and possible LLMjacking. Compare provider usage with your own calls, contain unauthorized spend and prepare refund evidence.+ - [Coding-agent cost tracking across tools and teams](https://veqen-app.com/guides/coding-agent-costs.md): Track Claude Code, Cursor, Copilot and other coding-agent costs per person and team. Combine seats, credits and tokens, check invoices and find idle seats.+ - [How to give an AI agent a company card safely](https://veqen-app.com/guides/give-an-ai-agent-a-card.md): What to set up before an AI agent spends on a company card or pays from an account, what a card platform's own limits catch and miss, and how to check every charge against what the agent was asked to do.+ - [Internal controls and audit evidence for AI agents that spend money](https://veqen-app.com/guides/controls-for-ai-agents-that-spend.md): What controllers and auditors will ask about AI agents that pay bills, hold cards or post entries, from COSO's 2026 guidance on generative AI to the AICPA and AIUC assurance engagement for agents, and the evidence to keep.+ - [AI cost per customer: track LLM usage and margin](https://veqen-app.com/guides/cost-per-customer.md): Attribute AI model calls to customers and features in Python or JavaScript. Calculate LLM cost per customer, check margins and set customer budgets.+ - [AI agent cost per task: measure every run](https://veqen-app.com/guides/ai-agent-cost-per-run.md): Calculate AI agent cost per completed task, including retries and failed runs. Attribute model calls, compare expensive runs and set useful spending limits.+ - [What to bring to an AI vendor renewal](https://veqen-app.com/guides/negotiate-ai-contracts.md): How to prepare for an OpenAI, Anthropic, Cursor or Copilot renewal: your effective rate against list price, commitment pacing and tiers, idle seats, and the notice date.+ - [How to prove an AI contract saved money](https://veqen-app.com/guides/prove-ai-vendor-savings.md): A before-and-after method CFOs, controllers and PE operating partners can accept: separate what a renegotiated AI rate saved from usage growth and the vendor's own price changes.+ - [What happens when you don't use your AI commitment](https://veqen-app.com/guides/ai-commitment-shortfall.md): Committed-spend AI contracts with OpenAI, Anthropic, Azure and cloud marketplaces: how a shortfall happens, how to see it coming, and what to do before the period ends.+ - [Stripe Issuing for AI agents](https://veqen-app.com/integrations/stripe-issuing.md): Watch the Stripe Issuing cards you give AI agents. Every authorization is matched to what the agent said it was doing, spend no action explains is caught within minutes, and the card can be frozen from the alert.+ - [Ramp agent cards](https://veqen-app.com/integrations/ramp.md): Watch the Ramp funds your agent cards spend from. Every transaction, declines included, is matched to what the agent said it was doing, and a fund can be suspended from the alert.+ - [Brex cards for AI agents](https://veqen-app.com/integrations/brex.md): Watch the Brex cards you give AI agents. Every posted transaction is matched to what the agent said it was doing, and the card can be locked from the alert.+ - [Mercury agent cards and payments](https://veqen-app.com/integrations/mercury.md): Watch Mercury cards, including Mercury's own agent cards, and the card spend, ACH and wire payments that leave the account your agents pay from, matched to what each agent said it was doing.

https://veqen-app.com/llms.txt

− > Veqen is a pay-as-you-go API and MCP server that checks whether a signup, email address, IP address, phone number or domain is real, throwaway or risky. Every result has a 0-100 risk score, a verdict (admit, review, refuse) and plain-English reasons.− - Auth: `Authorization: Bearer vq_live_...` (create a key at https://veqen-app.com/app/keys).− - Free tier: 1,000 credits a month. Credit costs: email 1, ip 1, phone 1, domain 1, signup 3, mailbox 5.− - OpenAPI: https://veqen-app.com/openapi.json− - MCP server (streamable HTTP, OAuth or API key): https://veqen-app.com/mcp− - Full docs in one file: https://veqen-app.com/llms-full.txt− - Open-source Python and JavaScript SDKs: https://github.com/Keyveve/veqen-sdks− - [Quickstart: your first check in 2 minutes](https://veqen-app.com/docs/index.md): Get an API key, make one request, and read the verdict. Copy-paste examples for curl, Python and JavaScript.− - [Authentication](https://veqen-app.com/docs/authentication.md): Send your API key in the Authorization header. How keys work, how to rotate them, and how to keep them safe.− - [Credits and billing](https://veqen-app.com/docs/credits.md): What each check costs, how free monthly credits work, prepaid packs, auto top-up, and what happens when you run out.− - [Errors](https://veqen-app.com/docs/errors.md): Every error has the same shape, a stable type and a plain-English message. Here's what each one means and how to fix it.− - [Rate limits](https://veqen-app.com/docs/rate-limits.md): How many requests per second you can send, the headers that tell you where you stand, and how to handle 429s.− - [Idempotency (safe retries)](https://veqen-app.com/docs/idempotency.md): Send an Idempotency-Key header and retries return the original answer without charging you twice.− - [Signup check](https://veqen-app.com/docs/signup-check.md): Email, IP and phone in one call, with cross-checks between them and a single verdict. The recommended way to screen new accounts.− - [Email check](https://veqen-app.com/docs/email-check.md): Check one email address for throwaway domains, typos, missing mail servers, new domains, role inboxes and (optionally) the mailbox itself.− - [Email list (batch)](https://veqen-app.com/docs/email-batch.md): Check up to 100 email addresses in one request. Same result per address as the single email check.− - [IP check](https://veqen-app.com/docs/ip-check.md): Find out where an IP address is, who owns its network, and whether it's a datacenter, VPN network, privacy relay or Tor exit.− - [Phone check](https://veqen-app.com/docs/phone-check.md): Validate a phone number, format it as E.164, and find its country, line type (mobile, VoIP, toll-free...) and original carrier.− - [Domain check](https://veqen-app.com/docs/domain-check.md): Is a company domain real? Whether it exists, receives email, is a throwaway provider, and when it was registered.− - [MCP server for AI agents](https://veqen-app.com/docs/mcp.md): Connect Veqen to Claude, ChatGPT, Cursor or any MCP client. OAuth or API key, the tools available, and what they cost.− - [SDKs and OpenAPI](https://veqen-app.com/docs/sdks.md): Official Python and JavaScript clients, the OpenAPI spec, and docs formatted for AI coding assistants.− - [Pay per call with x402](https://veqen-app.com/docs/agent-payments.md): Agents with a crypto wallet can pay for single checks in USDC over the x402 protocol, with no account or API key.− - [Risk scores and reasons](https://veqen-app.com/docs/reasons.md): How scores, levels and verdicts work, and every reason code we return with its impact and meaning.− - [Balance and usage](https://veqen-app.com/docs/account.md): Read your credit balance and daily usage from code. Both endpoints are free.− ## Free tools− - [Disposable email checker](https://veqen-app.com/tools/disposable-email-checker): Free disposable email checker. Paste an address or domain to see if it's a temporary, throwaway inbox (Mailinator, 10 Minute Mail, Guerrilla Mail and thousands more).− - [Email address checker](https://veqen-app.com/tools/email-checker): Free email address checker. Validates syntax, catches typos like gmial.com, checks the domain's mail servers, and flags disposable, role and brand-new-domain addresses.− - [IP lookup](https://veqen-app.com/tools/ip-lookup): Free IP address lookup. See the country, city and network owner of any IPv4 or IPv6 address, and whether it's a datacenter, VPN network, privacy relay or Tor exit node.− - [VPN & proxy detector](https://veqen-app.com/tools/vpn-detector): Free VPN and proxy detector. Find out if an IP address belongs to a datacenter, a VPN-heavy network, a privacy relay or a Tor exit node.− - [Phone number validator](https://veqen-app.com/tools/phone-validator): Free phone number validator for every country. Checks if a number is real, formats it as E.164, and shows its line type (mobile, landline, VoIP, toll-free) and original carrier.− - [Domain age checker](https://veqen-app.com/tools/domain-age-checker): Free domain age checker. See when a domain was registered, its registrar, whether it receives email, and whether it publishes SPF and DMARC.− - [MX lookup](https://veqen-app.com/tools/mx-lookup): Free MX lookup. List a domain's mail servers in priority order and check whether it publishes SPF and DMARC, or refuses mail with a null MX.− - [Signup risk checker](https://veqen-app.com/tools/signup-risk-checker): Free signup risk checker. Combine an email, IP address and phone number into one score and verdict (admit, review, refuse) with plain-English reasons.− ## Optional− - [How Veqen scores signup risk: sources, thresholds and limits](https://veqen-app.com/guides/how-veqen-scores-signup-risk): See how Veqen combines email, IP and phone signals, where its data comes from, and why a risk score is a screening suggestion rather than an identity check.− - [How to block disposable email addresses at signup](https://veqen-app.com/guides/block-disposable-emails): Why blocklists alone miss new throwaway domains, the three checks that catch them, and copy-paste code to block them in your signup form.− - [How to stop free trial abuse without hurting real signups](https://veqen-app.com/guides/stop-free-trial-abuse): The patterns behind repeat free-trial signups, which signals catch them, and a tiered response (admit, verify, refuse) that keeps conversion high.− - [Protect your AI app's free credits from abuse](https://veqen-app.com/guides/protect-ai-app-free-credits): Free AI credits are worth real money, so they attract scripted signups. How to stop credit farming without adding friction for real users.− - [How to detect VPN, proxy, datacenter and Tor traffic](https://veqen-app.com/guides/detect-vpn-proxy-tor): How IP intelligence tells a person on home broadband from a server, what each signal means, and how to respond without blocking real users.− - [How to validate phone numbers (and spot VoIP numbers)](https://veqen-app.com/guides/validate-phone-numbers): Parse what people type, store it in E.164, reject impossible numbers, and flag VoIP, premium-rate and toll-free lines, with code for every stack.+ > Veqen is cost monitoring and protection for AI. For work and coding agents, it shows each person's and team's cost across ChatGPT, Claude, Microsoft 365 Copilot, Claude Code, Cursor, GitHub Copilot, Codex, Devin and the other agents, priced at the company's contracts, with limits pushed into each tool, misuse alerts, idle seats and chargeback. For product inference, it reads each provider's billing and per-key usage to catch stolen keys and unexplained spend within the hour, shut the key off, + - Work and coding agents, per person: https://veqen-app.com/work-agents+ - Product inference and stolen-key protection: https://veqen-app.com/product-inference+ - The free audit of an AI bill: https://veqen-app.com/audit+ - Price book as JSON: https://veqen-app.com/prices.json (version 2026-10-28)+ - Current product release: Veqen 2.0.0; changes: https://veqen-app.com/docs/whats-new.md+ - Agent setup and safe tool selection: https://veqen-app.com/docs/agents.md+ - AI costs explained, with a worked estimate: https://veqen-app.com/guides/ai-costs.md+ - Model prices: https://veqen-app.com/models+ - Cost simulator for a whole workload: https://veqen-app.com/simulate (API: POST https://veqen-app.com/v1/simulate, no key)+ - API reference (nearly everything the dashboard does): https://veqen-app.com/docs/api.md; OpenAPI: https://veqen-app.com/openapi.json+ - MCP server (OAuth or a full-access key; reads spend and manages budgets, credits, bills, prices and month-end): https://veqen-app.com/mcp+ - Public pricing MCP server, no sign-in: https://veqen-app.com/mcp/prices+ - All docs in one file: https://veqen-app.com/llms-full.txt+ - [Veqen docs](https://veqen-app.com/docs/index.md): See who spent every AI dollar across work and coding agents, your product and your providers, catch stolen keys and spend nobody on your team made, and turn it into budgets, margin, chargeback and journal entries.+ - [Protection](https://veqen-app.com/docs/protection.md): How Veqen catches stolen keys and spend nobody on your team made, what each rule means, how to read an incident, shut a key off, file a refund claim, and get overcharges back.+ - [Quickstart](https://veqen-app.com/docs/quickstart.md): Wrap your AI client, tag calls with a customer and feature, and see the cost in under five minutes.+ - [Work and coding agents](https://veqen-app.com/docs/work-agents.md): Each person's and team's cost across every work and coding agent, from ChatGPT and Claude to Cursor and Devin, priced at your contracts, with limits inside the tools that take them, misuse alerts, idle seats, invoice checks and chargeback.+ - [How Veqen works](https://veqen-app.com/docs/how-it-works.md): The price book, tags, the local and cloud ledgers, and what happens to a call from your code to the dashboard.+ - [The free audit](https://veqen-app.com/docs/audit.md): Connect read-only access to your AI providers and your work and coding agents and get a report of the last 30 days within the hour, free at any spend size.+ - [What's new in Veqen 2.0](https://veqen-app.com/docs/whats-new.md): Veqen 2.0 is cost monitoring and protection for AI: work and coding agents per person across every tool, stolen-key and unexplained-spend protection with shut-off and refund claims, overcharge checks, the free audit, chargeback, new plans, and SDKs 1.4 and 1.2.+ - [Python SDK](https://veqen-app.com/docs/python.md): veqen for Python. Wrap OpenAI, Anthropic, Google and any OpenAI-compatible client, for text, embeddings, images and audio; tag calls, set budgets, record any call by hand and price models offline.+ - [JavaScript and TypeScript SDK](https://veqen-app.com/docs/javascript.md): veqen for Node.js. Wrap OpenAI, Anthropic and Google clients, tag calls with AsyncLocalStorage, set budgets and price models offline.+ - [Framework integrations](https://veqen-app.com/docs/integrations.md): Track costs from the Vercel AI SDK, LangChain, LlamaIndex, LiteLLM, Haystack and the OpenAI Agents SDK without wrapping each client.+ - [LangChain and LangGraph](https://veqen-app.com/docs/langchain.md): Track token usage and estimated model costs in LangChain and LangGraph, attribute calls to customers and features, and check Veqen budgets before execution.+ - [OpenTelemetry](https://veqen-app.com/docs/opentelemetry.md): Send GenAI spans from the Vercel AI SDK, OpenLLMetry, Logfire or the OpenAI Agents SDK straight to Veqen, with no SDK in your code.+ - [Tags](https://veqen-app.com/docs/tags.md): Attribute every AI call to a customer, feature, user, project and environment, so cost per customer, feature, team or project comes for free.+ - [CLI and GitHub Action](https://veqen-app.com/docs/cli.md): Find every model in your code with its price and retirement date, review the cost of model changes in pull requests, audit last month's bill, and turn coding tools' exports into a bill per team, all locally.+ - [Budgets and alerts](https://veqen-app.com/docs/budgets.md): Limits on AI spend per organization, customer, feature, user or project, with alerts at 50, 80 and 100 percent and SDK enforcement that stops calls before they are made.+ - [Runs and conversations](https://veqen-app.com/docs/runs.md): See what one agent run, conversation or batch job cost from start to finish, how run costs spread, and which runs cost the most.+ - [Anomalies, savings and forecasts](https://veqen-app.com/docs/insights.md): How Veqen spots a spike on a customer, user, feature or model the same day, explains why spend moved, finds ways to spend less, and forecasts the month.+ - [Reconciliation](https://veqen-app.com/docs/reconciliation.md): Compare what providers billed with the calls you tracked: imported daily or every six hours from OpenAI, Anthropic, AWS Bedrock, Google Cloud, Azure and OpenRouter, or from any provider's bill or billing export, GPU clouds included.+ - [Alerts, saved views and reports](https://veqen-app.com/docs/alerts.md): What Veqen alerts on, where alerts go (email, Slack, a signed webhook), how delivery is retried and logged, and how to save a Breakdown view and have it emailed weekly or monthly.+ - [Custom prices](https://veqen-app.com/docs/custom-prices.md): Price calls at your negotiated discount, your committed-use rate, or what your own GPUs cost, for any provider and any kind of model.+ - [Margin per customer](https://veqen-app.com/docs/margin.md): Connect a read-only Stripe key and see revenue, AI cost and margin for every customer, least profitable first.+ - [Provider credits](https://veqen-app.com/docs/credits.md): Track prepaid and promotional credits with OpenAI, Anthropic and other providers, see the drawdown, and get warned before they expire.+ - [Monthly close](https://veqen-app.com/docs/close.md): Classify AI spend by category, department, class and legal entity, book what providers billed, draw down credits, accrue the rest in your books' currency, true up late invoices, and export journal entries for QuickBooks Online and Xero.+ - [AI analyst](https://veqen-app.com/docs/ai-analyst.md): Ask about your AI spend in plain English, on the dashboard or through the API, and get a short answer, why, what to do, and the exact numbers behind it.+ - [MCP server](https://veqen-app.com/docs/mcp.md): Ask Claude, ChatGPT, Cursor or any MCP client about your AI spend, and have it run everything the dashboard does, or look up model prices with no sign-in.+ - [Organizations, members and keys](https://veqen-app.com/docs/teams.md): Sign-in, organizations, roles, invitations, API key scopes and the audit log.

Showing the first 40 of each.

description

Scale AI inference without losing track of the bill. See what costs most and where you could save. → AI cost monitoring and protection: spend per person and customer, stolen-key alerts and bill checks.

title

Veqen: the cost ledger for AI inference → Veqen: AI cost monitoring and protection

version

1.5.0 → 2.0.0

description

AI spend by customer, feature and model, reconciled to provider bills, with budgets and margin. → Scale AI inference without losing track of the bill. See what costs most and where you could save.

version

1.0.0 → 1.5.0

description

Screen signup risk using email, IP, phone and domain checks with scores and plain-English reasons. → AI spend by customer, feature and model, reconciled to provider bills, with budgets and margin.

title

Veqen signup checks → Veqen: the cost ledger for AI inference

version

0.1.0 → 1.0.0