Your app relies on a dozen metered APIs. Argus watches the balances, spend, and rate limits on all of them, and tells you before one runs dry.
Build updates as we ship, and your invite when a spot opens.
Consumption pricing put a meter on everything you build with — credits, tokens, calls, rate limits. The providers all bill flawlessly; none of them tell you where you stand. So you check ten dashboards, or you check none and find out when production stops.
We found out the second way. The week Argus started, three of our own meters were already dry.
Real numbers. Argus read all three in one poll.
Poll for authoritative totals, push for per-request detail, MCP for the agents doing the spending.
openrouter · sk-or-•••••••••••• → balance · usage · plan, hourly
Connect a provider and Argus polls its billing endpoints on your behalf: authoritative totals, remaining credits, plan status. No code.
globalThis.fetch = argus.instrument(fetch);
One line at your app's edge. Every call gets itemized as it happens: cost, latency, and the rate-limit headers billing pages never show.
"url": "https://api.argus.town/mcp" → 7 read-only meter tools
Agents spend faster than people. Over MCP, yours can check its own runway and steer around a provider that is about to cut it off.
Or hand it to your agent — paste this into Claude Code, Cursor, or any harness:
Tracks the whole stack






…and any other API you call.
Every balance gets a burn rate and a countdown. You hear about it at seven days, not at zero.
RPM, TPM, and 429s measured from your real traffic. See the wall before you hit it.
Runway short, balance low, spend spiking, rate limit closing in. One email, in time to act.
Not just LLMs. The enrichment, scraping, and data APIs nobody else meters.
A read-only MCP server, so the things doing the spending can watch the spending.
Provider credentials are AES-256 encrypted at rest, never logged, never returned by the API.
We're letting people in slowly. Leave an email and you're in line.