For AI-native teams

Your agents' reports, read outside the terminal.

Eval runs, benchmarks, release and incident reports from CI and headless agents, published as private reports in your brand. Execs, customers and auditors open them with a one-time email code. No agent-tool account needed.

Apply for the pilotTake the tour

Design-partner pilot open. Works with Claude, Codex, Cursor and any script or CI job that can call an API.

How it works

Set up your brand once. Every report after that is ready to read.

Follow one nightly eval report for Corvid Labs, a made-up AI product team, from brand kit to the people who decide.

  1. Step 1 of 3

    Add your brand

    Logo, colours and fonts, once. We check the colours stay readable in charts and tables.

    On the Team plan, reports open on your own domain.

  2. Step 2 of 3

    Connect your agents and jobs

    Give a CI job or headless agent an API key, or add QuickReeds to Claude, Codex or Cursor. The job publishes with one call.

    The agent sends data and words. Our renderer lays out the report, so the numbers on the page are the numbers it was given.

  3. Step 3 of 3

    Approve who reads it

    An agent can request a share, and a person approves it. Execs, customers and auditors open the report with a one-time email code. They don't need your agent tool.

    Nothing leaves your team without a person's approval.

Where reports end up

The work is done. Nobody outside the team ever reads it.

Stuck where it ran

Results land in CI logs and Markdown files in a repo, where only the people who know where to look will find them.

Only your tool can open it

A Claude artifact is easy to make and hard to share: only people who use Claude can open it. Execs, customers and auditors usually don't.

No safe way to share

No password links, no expiry, no own domain. Sending results outside the team means a public link or an export by hand.

Why we built it

The people who decide should be able to read what your agents found.

Your agents already run the evals, benchmarks and release checks. We think they should hand off to something that publishes the results in your brand, the same way every time, and that a person, not an agent, should decide who reads them.

Built on Cloudflare, with every customer's data separated at the database level.

Our promises

  • Your readers see your brand, not ours.
  • Nothing leaves your team without a person's approval.
  • The numbers on the page are the numbers in your data.

What changes

From work that disappears into CI logs to work read by the people who decide.

Without QuickReeds

  • Reports in CI logs, Markdown files and chat artifacts
  • Readers need the same agent tool to open them
  • A number doesn't match the data, and a customer spots it first
  • Confidential results on a public link that never expires

With QuickReeds

  • Execs, customers and auditors open reports with a one-time email code
  • Every report in your brand, laid out by our renderer
  • The numbers on the page are the numbers in your data
  • Nothing leaves the team without a person's approval

Planned pricing

Priced per team, with one brand included.

These are the prices we plan to launch with after the pilot. Pilot partners use QuickReeds free during the pilot, then get 50% off their first year.

Free

Trying it on your own

$0Planned

  • 50 live reports
  • Our default design with your logo
  • Reports on name.qckrds.com
  • Links with an expiry date
  • 3 named recipients per report
  • Small “Made with QuickReeds” badge

Pro

Freelancers and solo developers

$19/monthPlanned

  • Unlimited reports
  • Your full brand: logo, colours, fonts
  • No QuickReeds badge
  • Password-protected links
  • 25 named recipients per report
  • One year of version history

Team

For teams

Agencies, consultancies and teams

$49/monthPlanned

  • 5 members included, then $10/month each
  • 1 brand included, then $20/month per client brand
  • Each client brand gets its own domain and workspace
  • Your brand only, with 2+ brands: no QuickReeds on sign-in pages or emails
  • Unlimited named recipients
  • Roles and a full activity log
6

Team with 6 brands: $149/month, or $1,490/year with two months free.

Questions

Before you apply

Can I use QuickReeds today?

Not publicly yet. We're building it with a small group of design partners first. Pilot partners get access as soon as each part works, and we set up their brands for them.

Which AI tools does it work with?

Claude, Codex and Cursor connect through MCP. Anything else that can make an HTTP request, such as a script, a scheduled job or a CI pipeline, can publish with an API key.

Does AI write or change the numbers in my reports?

No. Your agent sends the data and the words; our renderer turns them into the report without any AI involved. The numbers on the page are the numbers it was given.

Do readers need an account?

No QuickReeds or agent-tool account. Named readers confirm their email with a one-time code, or you send a password link that expires.

Can a CI job publish without a person?

A job can publish a report privately to your team at any time. Sharing it with anyone outside the team needs a person's approval.

What if our reports need something your templates can't do?

Tell us during the pilot. That's exactly what the pilot is for.

Design-partner pilot

Pilot it with your eval, release or incident reports.

We're looking for a few engineering, data and ML teams whose agents or CI jobs produce reports that people outside the team need to read.

  • Free during the pilot, then 50% off your first year.
  • We set up your brand for you, and you get a direct line to the founder.
  • In return: a 20-minute call now, honest feedback during the pilot, and a short case study if it works for you.

We reply to every application within two business days.

What best describes your team?
Which AI tools does your team use?

We only use your details to contact you about the pilot. See our privacy notice.