Cycling through all six. Tap any point to stop.

Measured on six things

Third-party AI API access for coding workflows, with Claude Code configuration guides for VS Code, Cursor and JetBrains.

Security

HTTP probe found HTTPS not enforced, no HSTS, CSP, frame protection, Referrer-Policy, SPF or DMARC — 0/13 checklist points. No audit or compliance claims on site.

Screenshots

1 / 1

AI in IDE

About this app

AI in IDE is an independent third-party service for developers using AI models in coding workflows. Its website provides Claude Code installation and API configuration guides for VS Code, Cursor and JetBrains. Developers configure the API base URL and add their own AI in IDE key. The service offers paid subscription plans with usage quotas and a paid three-day trial. Pricing starts from $12.99/month; there is no free tier. It is not an official Anthropic service or an authorized reseller.

Key features

1One account for Claude, GPT and Gemini API quota via channel-based keys (AWSQ/CCMAX, Codex, Gemini)
2Setup guides for Claude Code in Cursor, JetBrains and VS Code, plus Codex and Gemini configuration
3Subscription tiers plus separate top-up balances ($50–$2,000) with configurable spending priority

Target customer profile

Drawn from the product itself, not from a survey.

Scout Review

Written by AI from measured evidence, scored out of 100.

AI in IDE is an independent API gateway that sells Claude, GPT and Gemini quota through one account, with setup guides for Claude Code, Cursor, VS Code and JetBrains. The site is fast (Lighthouse 94 mobile / 97 desktop) and the docs are more substantial than most indie tools manage. The problems are structural: a security probe that scores 0/13 with HTTPS not enforced, a pricing page that reads like a contract, and a business model that competes directly with the official vendor while disclaiming any authorization from it. It's a competent front end on a proposition that needs a lot more trust than it currently earns.

Measure by measure

Usability

Docs are real and cover Cursor, JetBrains, VS Code, Codex and Gemini, and the install snippet is a strong first step. But pricing terms (channels, quota multipliers, renewals) hide in fine print and accordions.

Design

Coherent warm palette and a genuinely useful terminal snippet in the hero, but the Double 11 promo repeats three times, badge colours clash, and secondary copy is low-contrast grey.

Speed

Lighthouse mobile 94/100 (LCP 1.1s, CLS 0, TBT 149ms) and desktop 97/100. Genuinely fast, no lag complaints in the collected evidence.

Security

HTTP probe found HTTPS not enforced, no HSTS, CSP, frame protection, Referrer-Policy, SPF or DMARC — 0/13 checklist points. No audit or compliance claims on site.

Accessibility

Lighthouse accessibility 96/100, with two real failures: insufficient colour contrast and visible labels without matching accessible names. Language switcher present; no WCAG statement found.

Growth

Real problem — multi-provider API access in one account — but the wedge is thin: an unofficial gateway competing with Anthropic's own billing, with no free tier to sample.

Improvement areas

Security headers absent across the board (0/13 probe): enforce HTTPS, add HSTS, CSP, SPF and DMARC
Pricing complexity buried in fine print and accordions; surface plain per-plan prices and quotas
No free tier and a paid 3-day trial make it hard to test gateway reliability before committing

Design & Usability

The homepage has a real point of view: warm neutrals, one rust accent, and a hero that shows the actual npm install and export lines before any marketing. That's the right instinct for a developer tool. The execution slips below the fold — the Double 11 promotion fires three times in one scroll, the channel cards use four unrelated badge colours, and the fine print is small light grey on cream. Usability is carried by the docs, which genuinely cover Cursor, JetBrains, VS Code, Codex and Gemini, and by the copy button on the install block. What drags it down is the pricing page: channels, quota multipliers, first-period discounts, renewal prices, expiring top-ups, and a spending-priority setting, all in accordions and small type. Four equal-weight CTAs in the pricing section don't help anyone decide.

Speed & Security

Speed is the strongest measured dimension: Lighthouse mobile 94/100 with LCP 1.1s, CLS 0 and TBT 149ms, desktop 97/100. Whatever else is true here, the site is snappy. Security is the opposite story. The HTTP probe returned 0/13: HTTPS not enforced, no HSTS, no CSP, no frame protection, no X-Content-Type-Options, no Referrer-Policy, no Permissions-Policy, no secure cookies, and no SPF or DMARC records. For a service whose entire job is holding API keys and metering paid quota, that's a serious gap, not a missing checkbox. The site is honest that it is an independent third-party gateway and not an official Anthropic service or authorized reseller, which is the right disclosure — but disclosure is not a security program.

Accessibility & Growth

Accessibility scores well on the automated audit at 96/100, but two failures are real and fixable: background/foreground contrast, and visible text labels that don't match their accessible names. The contrast failure lines up with the visual audit's note about small light-grey copy on cream, so it's a systemic typography choice rather than a one-off. There's an EN language switcher but no accessibility statement. On growth: the problem is real — developers juggling Claude, GPT and Gemini keys and billing want one account — and the channel model (AWSQ/CCMAX for Claude, Codex for GPT, Gemini channel) is a concrete mechanism, not a slogan. But the wedge is thin. The listed alternative is Anthropic's own Claude Code, billed through official subscriptions or API usage, and this product explicitly says it is not authorized by them. There's no free tier and the trial costs money, so there's no low-risk way to find out whether the gateway is reliable.

Conclusion

If you're going to route paid API traffic and keys through a third party, the trust bar is higher than for a to-do app. Right now the site asks for that trust while failing every basic transport and email-authentication check, and while burying the actual cost of a plan in accordions. The fix list is not exotic: enforce HTTPS, add HSTS and a CSP, publish SPF and DMARC, put a plain price table above the fine print, and offer a free trial tier so people can test reliability before paying. The docs and the speed show the team can build. The security and pricing work is what decides whether anyone should hand them a key.

Suggestions for improvement

Enforce HTTPS and add HSTS, CSP, X-Content-Type-Options, Referrer-Policy and Permissions-Policy headers.
Publish SPF and DMARC records for aiinide.com so transactional email can't be spoofed.
Put a plain per-plan price and quota table above the accordions; move renewal and top-up rules below it.
Offer a free trial tier or free credits so developers can test gateway reliability before paying.

How we score

Alternatives comparison

Named competitors, point by point. Nobody paid to appear here or to be left out.

	AI in IDE	Claude Code (Anthropic)
Official vendor authorization	Explicitly not an official Anthropic service or authorized reseller	First-party Anthropic product
Multi-provider access (Claude, GPT, Gemini)	Yes — Claude, GPT and Gemini channels on one account	Claude models only
Billing model	Subscriptions from $9.09–$12.99/mo plus top-ups; paid 3-day trial, no free tier	Claude subscription plans or metered API usage
Security posture	Probe 0/13: HTTPS not enforced, no HSTS/CSP/SPF/DMARC	Vendor-managed infrastructure and compliance program
Setup documentation	Guides for Cursor, JetBrains, VS Code, Codex and Gemini	Official Anthropic documentation and CLI reference