Credits and pricing
How cloud credits work, what an audit costs, and how to control spend
Cloud audits are priced in credits - small integers, not dollars. Local audits never cost credits.
Getting credits
| Source | Credits | Notes |
|---|---|---|
| Free plan | 500/month | Every account, no card required; resets monthly on your signup date |
| Pro plan (190/yr) | 3,000/month | Resets each billing period; unused monthly credits expire |
| Team plan (290/seat/yr) | 3,000/seat/month | Pooled across the org; see Team plan for seat billing rules |
| Credit top-up (min $10) | 100 per $1 | Paid plan only; pick any amount; never expire |
Annual billing is 12 months for the price of 10. Your monthly credits still arrive every month; you just pay once a year.
Monthly credits are spent before purchased top-up credits. Subscribe and top up from the Billing & Credits page in the dashboard.
Free, Pro, and Team run the same audit at the same price: the base charge includes every cloud service (AI analysis, editor’s summary, blocking check, tech + site metadata, dead-link checking, report publishing) on every plan. Scheduled audits are on every plan: free re-audits one website weekly or monthly, Pro and Team re-audit every website daily. Pro and Team just grant more monthly credits and unlock top-ups, that daily cadence and extra scheduled sites, custom headers, and higher render concurrency - Team adds pooled credits, invites, and roles on top.
Max pages per cloud audit
Audits triggered from the dashboard - on-demand (“Run now”) and scheduled - crawl up to a plan-based page ceiling:
| Plan | Max pages per cloud audit |
|---|---|
| Free | 500 |
| Pro | 2,000 |
| Team | 10,000 |
Free stays at 500 for a technical reason rather than a pricing one: free audits run on a smaller container, and a 10,000-page audit needs several gigabytes of memory to hold the crawl.
A crawl config above your plan’s ceiling isn’t an error - it’s clamped automatically, the same way render_concurrency is. This limit is separate from the CLI’s own crawl cap: squirrel audit run locally on your machine crawls up to 10,000 pages on every plan - see Crawl Limits.
If you’re driving audits through the REST API or the MCP server, a clamped run’s response carries a notice field (type: "page_limit_clamped", with the requested and applied page counts and an upgrade hint) so your integration or agent can surface it - the audit still runs normally either way.
Checking your balance
squirrel creditsPlan: Pro
Balance: 2340 credits (2000 monthly + 340 purchased)
monthly credits reset 2026-07-08
Pricing:
audit 50 base + 2 per rendered page
(analysis, tech detection, publishing included)
content_gaps 25 per run
issue_enrich 3 per issue
keyword_gaps 25 per run
Top up: https://squirrelscan.com/upgrade?src=cli-creditssquirrel auth status also shows your balance, and the start banner of every audit shows the estimated spend against it.
Pricing
Every authed audit has one flat price:
| Item | Cost | Notes |
|---|---|---|
| Audit base | 50 per audit | covers every included service below |
| Browser rendering | 2 per rendered page | Cloud rendering; charged whether or not the page hits the render cache |
The base includes everything: AI page analysis (ai/llm-parsability, ai/page-type-match), authority signals, the Blocking check, technology detection, site metadata, the editor’s summary, dead-link checking, and publishing the report. None of these are charged separately.
HTTP-only (non-rendered) page fetches are free, so a --no-render audit costs just the 50-credit base.
Opt-in add-ons are the only extra charges:
| Add-on | Cost | Notes |
|---|---|---|
| Keyword gaps | 25 per audit | gaps/keywords (opt-in) |
| Content gaps | 25 per audit | gaps/content (opt-in) |
| Issue enrichment | 3 per issue | AI recommendations in the dashboard |
| Dead-link render escalation | 2 per link | WAF-blocked links retried via a rendered fetch, billed as rendered requests |
A typical 25-page rendered audit costs 100 credits: 50 base + 50 for rendering. A 100-page audit costs 250 credits.
Spend controls
Three guardrails keep an audit from spending more than you expect:
[cloud]
max_credits_per_audit = 1000 # hard cap per audit; 0 = unlimited
confirm_threshold = 50 # estimates above this prompt before spending
enabled = true # master switch for all cloud features- Per-audit cap - if the estimate exceeds
max_credits_per_audit, work is truncated deterministically at the cap and remaining rules reportskippedwith reasoncredit-cap-reached. - Confirmation prompt - in a terminal, an estimate above
confirm_thresholdasks before proceeding. Use--yesto skip prompts (e.g. in CI or agent workflows). Non-interactive runs proceed without prompting, bounded by the cap. - Master switch -
enabled = falseturns off all cloud calls for the project.
See the cloud configuration reference for all options.
When credits run out
Nothing breaks. A cloud audit needs a balance of at least 50 credits (the base) to start: below that, the CLI runs the audit locally instead and tells you so. If credits run out mid-audit, cloud rules report skipped with reason insufficient-credits, the audit completes with all local rules, and you see a single warning with a top-up link. If an audit fails after you were charged, the charge is automatically refunded.