# Argus > Argus lets your coding agent track API spend while you build. Over MCP it can check what a run cost, catch a loop burning credits, and see a provider running dry. You get one place for every metered API's balance, spend and runway, with alerts before one hits zero. Argus lets your coding agent keep track of your API spend while you develop. Connect it to Claude Code, Cursor, Codex or any MCP client and the agent can check what a run cost, spot a leak (a loop burning credits, a forgotten cron), notice a provider running dry, and see which calls are worth optimizing. Underneath is one ledger for every metered API you pay for: balances, daily spend, plan quotas, burn rate, days to zero and rate-limit headroom, polled hourly from each provider's own billing endpoints and itemized per call by a fetch wrapper, with email alerts before a balance hits zero. Argus is not an LLM tracing tool; it meters spend across providers. ## How it works - [MCP](https://argus.town/docs/mcp): https://api.argus.town/mcp. Read tools get_overview, get_request_costs, get_advisories, get_rate_limit_status, get_cost_accuracy; management tools connect_provider and sync_connection. - [Poll](https://argus.town/integrations): paste a provider key; Argus reads balances, daily spend and plan quotas every hour. - [SDK](https://argus.town/docs/sdk) and [ingest](https://argus.town/docs/ingest): wrap fetch once to itemize every call with cost, tokens and rate-limit headers, for any provider. ## Integrations: what each provider's API exposes - [Anthropic](https://argus.town/integrations/anthropic): API exposes daily spend, per-model cost. - [Apify](https://argus.town/integrations/apify): API exposes balance, plan. - [Apollo.io](https://argus.town/integrations/apollo): API exposes balance. - [Bright Data](https://argus.town/integrations/bright-data): API exposes balance. - [Coresignal](https://argus.town/integrations/coresignal): API exposes balance. - [Crunchbase](https://argus.town/integrations/crunchbase): no billing API; Argus derives cost from pricing. - [Crustdata](https://argus.town/integrations/crustdata): API exposes balance, plan. - [DeepSeek](https://argus.town/integrations/deepseek): API exposes balance. - [Exa](https://argus.town/integrations/exa): no billing API; cost is reported on every response. - [Firecrawl](https://argus.town/integrations/firecrawl): API exposes balance, plan. - [Google Gemini](https://argus.town/integrations/google-gemini): no billing API; Argus derives cost from pricing. - [Google Places](https://argus.town/integrations/google-places): no billing API; Argus derives cost from pricing. - [HubSpot Breeze](https://argus.town/integrations/hubspot-breeze): no billing API; Argus derives cost from pricing. - [Hunter.io](https://argus.town/integrations/hunter): API exposes balance, plan. - [Instantly](https://argus.town/integrations/instantly): API exposes balance, plan. - [LeadMagic](https://argus.town/integrations/leadmagic): API exposes balance. - [MillionVerifier](https://argus.town/integrations/millionverifier): API exposes balance. - [OpenAI](https://argus.town/integrations/openai): API exposes daily spend, per-project cost, plan. - [OpenRouter](https://argus.town/integrations/openrouter): API exposes balance, plan. - [Parallel](https://argus.town/integrations/parallel): no billing API; Argus derives cost from pricing. - [People Data Labs](https://argus.town/integrations/people-data-labs): API exposes balance. - [Perplexity](https://argus.town/integrations/perplexity): no billing API; cost is reported on every response. - [Postmark](https://argus.town/integrations/postmark): no billing API; Argus derives cost from pricing. - [Prospeo](https://argus.town/integrations/prospeo): API exposes balance. - [SecurityTrails](https://argus.town/integrations/securitytrails): API exposes balance, plan. - [SendGrid](https://argus.town/integrations/sendgrid): API exposes plan. - [Tavily](https://argus.town/integrations/tavily): API exposes balance, plan. - [WhoisXML API](https://argus.town/integrations/whoisxml): API exposes balance. ## Pricing Flat monthly plans keyed to tracked API spend, never a percentage of it. Limits are soft (Argus keeps working past them and emails at 80% and 100%), except AI connector builds, which are a hard monthly count. - Free: $0. Up to $1,000 tracked spend/mo, 100K events/mo, 30-day history, 1 member, no AI connector builds. - Pro: $39/mo ($390/yr). Up to $15,000 tracked spend/mo, 3M events/mo, 90-day history, 5 members, 2 builds/mo. - Team: $149/mo ($1,490/yr). Up to $100,000 tracked spend/mo, 20M events/mo, 1-year history, unlimited members, 10 builds/mo. - Above $100,000/mo: custom pricing, by conversation. ## Guides - [Let your coding agent track API spend with MCP](https://argus.town/guides/coding-agent-api-spend-mcp): Give Claude Code, Cursor or Codex a read-only view of every metered API you build with, so it can check what a run cost, spot a leak, and notice a provider running dry. - [Track API spend across providers in one place](https://argus.town/guides/track-api-spend-across-providers): What each provider actually exposes over its API, how to put balances, daily spend and plan quotas on one ledger, and where you'll have to estimate. - [Find what's burning your API credits](https://argus.town/guides/find-what-is-burning-api-credits): A runaway loop, a forgotten cron, a retry storm: how to find the call that's draining a balance, from the burn rate down to the endpoint. - [Set a low-balance alert for OpenAI, Anthropic and OpenRouter](https://argus.town/guides/low-balance-alerts-openai-anthropic-openrouter): What each console's own alerts do, what they can't see, and how to alert on days of runway left instead of a dollar threshold. - [Check API credit balance programmatically](https://argus.town/guides/check-api-credit-balance-programmatically): The balance endpoint, auth header and response field for every major metered API, and the ones that only show a balance in the console. - [Estimate runway on prepaid API credits](https://argus.town/guides/estimate-runway-prepaid-api-credits): Days to zero is balance over burn rate. The hard part is the burn rate: top-ups, daily buckets and credits that aren't dollars. - [Why provider dashboards disagree with your bill](https://argus.town/guides/provider-dashboards-disagree-with-bill): Daily buckets, async settlement, cache pricing, per-request fees and charges that only bill on a match. Where the numbers drift, and how to reconcile them. - [What happens when API credits run out](https://argus.town/guides/api-credits-run-out-errors): The status code and error each provider returns at zero balance. Several look like rate limits or bad keys, so they get retried instead of fixed. - [Read rate-limit headroom from API response headers](https://argus.town/guides/rate-limit-headroom-headers): Which providers send remaining-requests headers, which send credit headers, and which send nothing, so you know before the 429. ## Comparisons - [Argus vs provider billing dashboards](https://argus.town/compare/provider-dashboards): OpenAI, Anthropic and OpenRouter consoles are authoritative and free. Where they're enough, and where checking ten of them stops working. - [Argus vs a cron job and a spreadsheet](https://argus.town/compare/spreadsheet): Polling balance endpoints yourself is a weekend project. Here's the honest list of what it takes and what you end up maintaining. - [Argus vs Ramp AI token spend management](https://argus.town/compare/ramp): Ramp reads OpenAI, Anthropic and other AI vendors' admin APIs daily for finance teams. Argus reads every metered API hourly for the developers and agents spending it. - [Argus vs Vantage](https://argus.town/compare/vantage): Vantage puts OpenAI, Anthropic and OpenRouter costs next to your cloud bill. Argus watches balances, runway and rate limits across the long tail of APIs. - [Argus vs CloudZero and Finout](https://argus.town/compare/cloudzero-finout): Enterprise FinOps platforms that ingest OpenAI and Anthropic costs for allocation and unit economics. When you need one, and when you need a meter. ## Docs - [API docs](https://argus.town/docs): first request in one curl. - [Authentication](https://argus.town/docs/authentication): API keys (read) vs sources (write). - [Query API](https://argus.town/docs/api): overview, connections, itemized events, providers, alerts, billing. - [Sending telemetry](https://argus.town/docs/ingest): the signed ingest POST, full event field reference. - [SDK](https://argus.town/docs/sdk): wrap fetch once; zero dependencies. - [MCP](https://argus.town/docs/mcp): meter tools for agents. - [For agents](https://argus.town/docs/agents): copyable prompt, Claude Code skill, AGENTS.md command. - [Full brief](https://argus.town/llms-full.txt): positioning, provider facts and the integration reference in one document. - [Claude Code skill](https://argus.town/skills/argus.md): ready-to-install SKILL.md. - [Pricing](https://argus.town/pricing) - [Changelog](https://argus.town/changelog) ## Status Argus is open. Sign up free at https://argus.town/signup.