Continuum Scan · free, no account

You have no idea what your AI coding costs.

Your CLIs show you a usage bar. They do not tell you which repo burned the money, which model was overkill, or the one thing that decides whether any of it matters: how much is prepaid on a subscription and how much is a metered API bill. One command reads the history already on your machine and tells you.

continuum scan
# reads local history, prints the report, sends nothing on its own $ curl -fsSL https://continuumcode.ai/scan.sh | sh
macos · linux · x64 · arm64 · ~7mb · no deps

Want to read it first? Inspect the script or run --dry-run to print the exact payload before anything is sent.

01 · The output

It prints a bill you were never shown.

Every number comes from the session files on your own disk, priced per model with the current API-equivalent rate card. No estimates from an industry average, no survey data, no telemetry we collected earlier.

continuum-scan - sample output
CONTINUUM AI spend report last 30 days · 1,284 sessions ── headline ───────────────────────────────────────────────── $4,812 of AI tokens burned in 30 days ($4,812/mo at this rate) $1,140/mo looks recoverable ($13,680/yr, 24% of spend) $172/mo of that is actual cash ($2,064/yr off a real bill) ── how you pay for it ─────────────────────────────────────── claude Claude Max 20x subscription $200.00/mo codex ChatGPT Pro subscription $200.00/mo openrouter openrouter api key price not detected on a subscription $4,468 prepaid - costs you nothing extra on metered API keys $344 billed per token to a card you pay $400.00/mo across 2 plans you got back 11x that in token value ── where it went ──────────────────────────────────────────── payments-api █████████████··· $2,918 61% web-dashboard ████··········· $968 20% infra-terraform ██············· $511 11% ── findings ───────────────────────────────────────────────── Context is being rebuilt instead of reused $604/mo Claude wrote 812M cache tokens against only 1.2B cache reads (0.68:1). A warm agent loop lands near 0.20:1. · 572M excess cache-creation tokens · Priced on claude-opus-5: $7,150 as writes vs $572 as reads Keep sessions resident instead of restarting per task. confidence: medium claude-opus-5 is doing work claude-sonnet-5 could handle $390/mo Averaging only 340 output tokens per request at 6.1x the price of a model already in your rotation. Route by task difficulty, not one model per session. confidence: medium

Illustrative figures. Your report is computed entirely on your machine.

02 · Subscription or API

A big number means nothing until you know who is billing you.

On a flat plan, tokens are prepaid: one more turn costs nothing, and "waste" is capacity you already bought. On an API key, every token is charged to a card. Same tokens, completely different decision. The scan reads how each CLI is signed in and splits your spend across both.

What it detects

Claude Pro, Max 5x and Max 20x. ChatGPT Plus, Pro and Team. Gemini, Cursor, Grok and the OpenCode connectors. It reads the plan tier, not just the fact that you are signed in, because Max 5x and Max 20x differ by $100 a month.

What that buys you

Your real return on the plans you pay for, as a multiple. Whether a subscription is idle enough that cancelling it is free money. And whether metered API spend is quietly running alongside a plan that already covers the work.

What it never reads

Enum fields only: the plan tier, the auth mode, whether a key is present. No token, key, refresh token, account id, or email address is ever read into the report, and the server drops them again on the way in.

Both credentials configured for one provider? It says mixed and leaves that spend unattributed, rather than guessing which one served a turn.

03 · What it looks for

Eight ways agentic coding leaks money.

Each finding names the observed quantity behind it, prices the counterfactual with the same rate card as the real spend, and states its own confidence. You can check the arithmetic, which is the point.

Context rebuilt, not reused

Cache writes cost roughly 12x what a cache read costs. Restarting a session per task pays that premium every time. The scan measures your write:read ratio and prices the gap.

Premium model on routine work

It measures the average turn size on each model. Small turns on your most expensive model are routine work paying frontier prices, and it names the next tier down that you already use.

Reasoning burn

Reasoning tokens bill at the output rate. An effort dial left at maximum for routine turns is charged in full, and shows up as a thinking-to-answering ratio well above the norm.

Correction loops

Runs of prompts that tell the agent it got it wrong, rather than moving the task forward. Every turn inside one re-pays for work already bought once.

Abandoned context

Sessions that opened, loaded the repo, and never got a second prompt. Sub-agent threads and scripted one-shot runs are excluded, because those are one prompt by design.

A plan that is not paying for itself

A subscription whose token value came in under its own monthly fee. This is the rare finding that is real cash: cancel it, or move work onto it until it earns out.

Metered spend beside a plan

API-key spend running alongside a subscription that already covers the work. Every one of those tokens is charged to a card while prepaid capacity sits unused.

Unpriced model sprawl

Models running with no published rate. That spend appears in no total anywhere, including this report's, which makes it the one number worth chasing down first.

04 · What leaves your machine

Nothing, until you ask.

The scan runs locally and prints the whole report to your terminal. It only offers to email you a copy afterwards, and it shows you what that copy contains before you decide.

Run it with --dry-run and it prints the exact JSON it would send, then exits. Run it with --local-only and it never asks at all.

never sent

Your prompts. The agent's replies. Any file contents, diffs, or code. File paths. Repository contents. Your username, hostname, or machine identifier. No API key, OAuth token, refresh token, account id, or email from any credential file.

sent, if you give an email

The aggregate numbers already printed in your terminal: totals, per-model and per-repo dollar figures, the findings, your detected plan names (Claude Max 20x) and billing rails, and bare repository folder names. Pass --anonymize-repos and even those become repo-1, repo-2.

05 · Then what

The findings point at an IDE.

Continuum is what fixes them: one warm session per repo so context survives between turns, per-turn model routing instead of one model per session, and a live spend meter across Claude, Codex, Gemini, Cursor, and Grok in one place.

The scan is free and standalone. You do not need Continuum to run it or to read the report.