All your teams’ AI spend. Under control.

Every team needs AI, but their monthly bills can bite. Get your team to use MindsHub - with all the models and tools, plus visibility and control over their spending.

Always-free usage for every team member. No trial period.

See how MindsHub works

Demo starts in 5 seconds.

Score 0

Misses 0 / 3

Space or tap to jump. Tap to jump.

The boss caught up.

Your score: 0

MindsHub

Give your teams shared access and spending controls.

Always-free usage for every team member. No trial period.

Try MindsHub

Interactive demo

See how MindsHub works.

Skip the demo

MindsHub

Welcome to Acme Labs

Remaining Balance$2,340.00

Free usageIncluded

The MindsHub Air model stays free within fair-use limits. Your balance covers paid models.

MindsHub Agents

Delegate a complete task

Research, analyze, create reports, build apps, or schedule recurring work.

MindsHub Inference

Use models in your code

Call models from your favorite agents with one key and one bill.

curl https://api.mindshub.ai/v1/chat/completions
  -H "Authorization: Bearer $MINDSHUB_KEY"
  -H "Content-Type: application/json"
  -d '{"model":"mindshub_air",
       "messages":[{"role":"user",
                    "content":"Hello!"}]}'

Organization / Billing

These records belong to your organization. Your role controls which billing details and actions you can access.

PlanPay as you go

Payment methodVisa ···· 4242

Remaining Balance $2,340.00

Usage Sep 1 – Oct 1, 2026

Credit spend this period

$4,040.00

Claude Sonnet 5.5 31% · $1,240.50
GPT-6 Astra 24% · $968.20
Claude Opus 5.5 17% · $706.00
Kimi K3 13% · $512.40
Gemini 3.8 Flash 10% · $398.75
GLM 5.3 5% · $214.15
MindsHub Air model 0% · $0.00

support-bot 28% · $1,140.00
Priya Nair 27% · $1,105.00
Sam Okafor 16% · $655.00
Omar Haddad 13% · $512.00
Lena Brooks 11% · $428.00
Dana Reyes 5% · $200.00

Engineering 49% · $1,960.00
Support 28% · $1,140.00
Data 13% · $512.00
Marketing 11% · $428.00

support-bot-prod 28% · $1,140.00
hub:anton:7c2048e1d3 17% · $705.00
sam-research 16% · $655.00
eval-pipeline 13% · $512.00
campaign-drafts 11% · $428.00
claude-code-dev 10% · $400.00
dana-notebook 5% · $200.00

Free allowance

Dana Reyes · Your usage

5-hour usage22% used

Resets in 3h 42m

7 budgets 1 at the limit 2 over 80%

Search budgets Applies to: Any Model: Any Status: Any

Budget	Spend	Used	Status
	of	0%	Normal
Acme LabsOrganization • All models	$4,040.00 of $8,000 / month	51%	Normal
EngineeringTeam • Claude Opus 5.5	$706.00 of $800 / month	88%	Over 80%
SupportTeam • All models	$1,140.00 of $1,500 / month	76%	Normal
DataTeam • Kimi K3	$512.40 of $500 / month	102%	Limit reached
Priya NairMember • Claude Sonnet 5.5	$382.10 of $400 / month	96%	Over 80%
support-bot-prodAPI key • All models	$41.20 of $75 / day	55%	Normal
DefaultWorkspace • All models	$3,612.00 of $6,000 / month	60%	Normal

Organization / Members

Organization roles

Organization roles apply across the organization. Workspace access also depends on workspace grants.

OwnerIncludes every Admin permission and protected ownership controls.

AdminManages members, invitations, teams, API keys, invoices, and billing (top-ups, card, auto-recharge), on top of everything a member can do.

MemberUses the product, sees who else is in the organization, and manages their own API keys.

User	Built-in role	Organization permissions
Dana Reyes (You)dana@acme.example	Owner	Built-in permissions
Priya Nairpriya@acme.example	Admin	Built-in permissions
Theo Granttheo@acme.example	Member	Member defaults
Sam Okaforsam@acme.example	Member	Member defaults
Lena Brookslena@acme.example	Member	Member defaults
Omar Haddadomar@acme.example	Member	Member defaults
Ana Silvaana@acme.example	Member	Viewer

Organization / Roles & permissions

These roles apply across the whole organization. Access inside a single workspace is decided on that workspace’s own Roles & permissions page. Compare the roles below, or create a custom role with the permissions your team needs.

Roles you can assign Owner and Admin hold the same permissions. What an Owner has that an Admin does not is ownership itself.

Owner Your role

Includes every Admin permission and protected ownership controls.

29 supported permissions

Cannot be given or taken away here

Admin

Manages members, invitations, teams, API keys, invoices, and billing (top-ups, card, auto-recharge), on top of everything a member can do.

29 supported permissions

Member

Uses the product, sees who else is in the organization, and manages their own API keys.

12 supported permissions

Viewer

A role we ship. Assigning it replaces the Member defaults.

10 supported permissions

Custom roles Roles somebody in this organization built. Assign one to a Member to replace their organization permissions.

Nobody here has built a role yet. The roles above are the ones we ship. Copy one to start from its permissions.

Showing 4 of 12 resource groups

Permission	Owner	Admin	Member
View members See who belongs to the organization. Read
Manage members Remove eligible organization members. Owners remain protected. Manage
Manage roles Create and assign custom roles, change built-in roles, and invite Admins. Manage roles

Permission	Owner	Admin	Member
View model rules See which models and providers are blocked for the organization, its workspaces and teams. Read
Manage model rules Block or allow a model or a provider for the organization, a workspace or a team. Manage

Permission	Owner	Admin	Member
Create API keys Mint a new API key for the caller’s own use. Create
Manage API keys Revoke any member’s API key. Only its owner can rename it. Manage
View API keys See the organization’s API key inventory. Read
Rename own API keys Rename keys belonging to you. Rename own keys
Revoke own API keys Revoke keys belonging to you. Revoke own keys

Permission	Owner	Admin	Member
View billing See the wallet balance, spend, and usage summary. Read
View invoices List and download the organization’s invoices. Read invoices
Top up wallet Add funds to the organization’s wallet and check a top-up’s status. Buy credits
Manage payment methods Add or replace the organization’s saved payment method. Manage payment methods
Manage auto-recharge Change automatic funding settings or resume paused automatic charges. Manage auto-recharge
Dismiss billing alerts Dismiss the organization’s low-balance warning. Dismiss alerts

Organization / Teams

Teams group organization members. Give a team an Admin or Member role in each workspace from Workspace access.

Name	Created on	Members	Workspace access	Role
All membersAll members · managed automatically	Aug 21, 2026	7 members	Workspace access	—
Engineering	Aug 24, 2026	4 members	Workspace access	Member
Marketing	Aug 24, 2026	1 member	Workspace access	Member
Support	Sep 2, 2026	1 member	Workspace access	Member
Data	Sep 9, 2026	1 member	Workspace access	Admin

Organization / Workspaces

Open a workspace to manage its settings and access. Organization roles and workspace roles are separate.

Name	Your workspace access	Created on
DefaultDefault · where unassigned resources live	Admin	Aug 21, 2026
Support bots	Admin	Sep 2, 2026

Organization / Model access

Each column shows that container’s own rules. Click a cell to block, allow, or clear one. A block anywhere wins, so a person in two containers is blocked if either blocks them. Until you add a rule, everyone can use everything.

Model	Everyone	Default	All members	Engineering	Marketing

Claude Sonnet
Claude Sonnet 5.5
Claude Sonnet 5
Claude Opus
Claude Opus 5.5
Claude Opus 5
Claude Fable
Claude Fable 5.1
Claude Fable 5
Claude Haiku
Claude Haiku 4.5

GPT
GPT-6 Astra
GPT 5.6 Sol
GPT Sol
GPT-6 Sol
GPT-6.1 Sol
GPT Terra
GPT 5.6 Terra
GPT Luna
GPT 5.6 Luna
GPT-6 Luna
GPT Codex
GPT 5.3 Codex
GPT Mini
GPT 5.4 Mini
GPT Nano
GPT 5.4 Nano
Text Embedding
Text Embedding 3 (small)

Gemini Pro
Gemini 3.1 Pro Preview
Gemini Flash
Gemini 3.8 Flash
Gemini 3.7 Flash
Gemini 3.6 Flash
Gemini 3.5 Flash
Gemini 3 Flash Preview
Gemini Flash-Lite
Gemini 3.1 Flash-Lite

Grok
Grok 4.7
Grok 4.6
Grok 4.5

Muse Spark
Muse Spark 1.3
Muse Spark 1.2
Muse Spark 1.1

GPT-OSS 120B (Fireworks)
Kimi
Kimi K3
DeepSeek Flash
DeepSeek V4.1 Flash
Qwen
Qwen3.8-2.4T-A95B
GLM
GLM 5.3
GLM Flash
GLM 5.3 Flash

MindsHub Blaze

Jev
Jev 1.13.0

MindsHub Air Always available

allowed by a rule blocked by a rule inherited no rule (allowed) model rules under this provider

Organization / Audit log

Who changed what in this organization, with what credential, and when. Shows the last 90 days; use the date filters below to narrow the range.

Filter by actor email Filter by action All objects After Before

When	Actor	Action	Object
10/2/2026, 3:17:10 PM	priya@acme.example	update	entitlement_policy	Details
10/2/2026, 3:17:01 PM	priya@acme.example	insert	workspace_grant	Details
10/2/2026, 12:25:12 PM	dana@acme.example	update	organization_member	Details
10/1/2026, 6:02:44 PM	dana@acme.example	insert	team	Details
10/1/2026, 9:40:03 AM	theo@acme.example	revoke	api_key	Details
9/30/2026, 4:11:29 PM	sam@acme.example	insert	api_key	Details
9/30/2026, 11:58:17 AM	priya@acme.example	insert	entitlement_policy	Details
9/29/2026, 2:20:51 PM	dana@acme.example	update	organization	Details

API Keys

Create and manage API keys for external use.

New keys belong to Acme Labs

Service Accounts only

Name	Key Prefix	Principal	Created	Status
support-bot-prod	mdb_8fQ2kL1x	Service Account	Feb 3, 2026	Active Revoked
hub:anton:7c2048e1d3Agent key	mdb_UVEVZZbt	priya@acme.example	Sep 18, 2026	Active Revoked
sam-research	mdb_XGlUqr53	sam@acme.example	Aug 29, 2026	Active Revoked
eval-pipeline	mdb_tdMUI2U7	omar@acme.example	Jul 9, 2026	Active Revoked
campaign-drafts	mdb_KWkOMJds	lena@acme.example	Sep 12, 2026	Active Revoked
claude-code-dev	mdb_gRYVNbHL	priya@acme.example	Sep 1, 2026	Active Revoked
dana-notebook	mdb_Pw3sHq7v	dana@acme.example	Aug 22, 2026	Active Revoked
openclaw-test	mdb_IUJKFXzf	theo@acme.example	Aug 30, 2026	Active Revoked

Traces

7dLast 7 days

Every inference request your account has made, and the conversations they belong to.

5,036 requests·0.4% error rate·278ms / 2.5s p50 / p95·1.0s / 12s streamed

Time	Request	Model	Latency	Status	Trace
Oct 2, 1:13 PM	—	sonnet-5-5	801ms	OK 200	cfef7ca3
Oct 2, 1:13 PM	—	sonnet-5-5	17s	OK 200	339843cc
Oct 2, 1:13 PM	—	mindshub_air	642ms	OK 200	c792986c
Oct 2, 1:13 PM	—	gpt-6-astra	716ms	OK 200	6f7db9dd
Oct 2, 1:13 PM	—	qwen-3-8-a95b	1.1s	OK 200	710e3797
Oct 2, 1:12 PM	—	opus-5-5	1.4s	OK 200	a3ba4be6
Oct 2, 1:12 PM	—	opus-5-5	37s	OK 200	ee274284
Oct 2, 1:12 PM	—	gemini-flash	2.8s	Failed 503	0ecee1d5
Oct 2, 1:12 PM	—	kimi	883ms	OK 200	a5f89255
Oct 2, 1:11 PM	—	sonnet-5	1.3s	OK 200	7c161182

Credit spend this period

$4,040.00

Total tokens in 1,654,169,609 incl. 669M cached · 2.9M long context

Total tokens out 145,719,745

Total web searches 214

Usage type	Tokens	Charge
Input	938M	$2,123.62
Cache write	44.6M	$136.26
Cache read	669M	$147.94
Output	146M	$1,574.09
Long context	2.9M	$58.09
Total		$4,040.00

Model	Input	Cache write	Cache read	Output	Long context	Charge
Claude Sonnet 5.5	329M ($657.47)	34.7M ($86.84)	248M ($49.62)	44.7M ($446.57)	0	$1,240.50
GPT-6 Astra	48.4M ($484.10)	0	29.1M ($29.05)	7.9M ($396.96)	2.9M ($58.09)	$968.20
Claude Opus 5.5	91.8M ($367.12)	9.9M ($49.42)	141M ($28.24)	13.1M ($261.22)	0	$706.00
Kimi K3	93.9M ($281.82)	0	68.3M ($20.50)	14M ($210.08)	0	$512.40
Gemini 3.8 Flash	287M ($215.33)	0	150M ($11.96)	45.7M ($171.46)	0	$398.75
GLM 5.3	84.1M ($117.78)	0	33M ($8.57)	20M ($87.80)	0	$214.15
MindsHub Air model	3.4M	0	0	380K	0	$0.00
Total	938M ($2,123.62)	44.6M ($136.26)	669M ($147.94)	146M ($1,574.09)	2.9M ($58.09)	$4,040.00

Code tasks

Recent 4

Agent

Recent tasks

Desktop app

Let's knock something off your list

general

Marketing

Last Google Ads campaign

Report on the performance of the last Google ads campaign.

Reviewed campaign performanceSynthetic sample campaign
Compared spend and leads by channel$4,800 spend · 160 leads
Created campaign-report.htmlInteractive report ready for review

The campaign generated 160 leads at $30 per lead. Brand search was the most efficient channel. Open the report to review the results with your team.

Q2-Reporting

Build an interactive KPI app the team can explore

Build me an interactive app the team can use to explore our KPIs — MRR, signups, NRR — pulling live from Postgres and Stripe. If something looks off with signups, let us open a Linear ticket and assign it right from the app. Publish it so the team can dig in.

PgPostgreSQLSStripeLLinear

Planned the analysis3 KPIs · live refresh · drill-down by segment
Connected to PostgreSQLvia the credentials vault — no raw key exposed
Queried Stripe for MRR & signups1,204 subscriptions reconciled
Wired up Linear for one-click assignmentissues created straight from flagged rows
Published kpi-app.htmllive, shareable URL

Done — here's your KPI app. MRR is up +9.4% week-over-week to $128.4k. One thing worth a look: signup checkout is erroring for EU accounts — I flagged it below so someone can grab it.

MRR, signups, and NRR update live — nothing to re-run.
Flagged a signup issue for EU checkout — assign it to a teammate right from the app.
Published to a live URL — the team can explore and assign follow-ups without asking you first.

Revenue-Ops

Top customer-feedback themes, ranked by revenue

Combine last month's Gong calls, Zendesk tickets, and HubSpot notes into the top customer-feedback themes — and rank them by revenue impact, not just frequency.

GGongZZendeskHHubSpot

Pulled 42 Gong call transcriptsMay 1 – May 31
Read 318 Zendesk ticketsplus 96 HubSpot deal notes
Clustered into 7 themessemantic grouping + dedupe
Scored each theme by attached ARRcross-referenced HubSpot deals
Wrote the memofeedback-themes.md

Here's the May feedback memo. Ranked by revenue, the #1 theme is onboarding friction — 28 mentions across $214k of ARR, well ahead of where raw frequency would rank it. Pricing confusion and a missing Salesforce sync round out the top three.

Every theme links back to the exact calls, tickets, and quotes.
I drafted a short product follow-up list at the end of the memo.

general

Extract 38 receipts into a Q2 expense workbook

Extract these 38 receipts from the shared Drive folder and build a categorized Q2 expense workbook with monthly totals. Flag anything that looks like a duplicate.

△Google Drive

Read 38 receipts from Google DriveOCR on PDFs and photos
Normalized vendors & currencies3 currencies → USD
Categorized into 12 bucketssoftware, travel, meals, …
Flagged 2 likely duplicatessame vendor, amount, ±1 day
Built expenses.xlsx3 sheets, formulas intact

Built expenses.xlsx — 38 receipts totaling $24,318 across 12 categories. Software is the biggest line at $9,240. I flagged 2 duplicate charges ($318) for your review on the Summary tab.

Each row links to the original receipt in Drive.
Monthly totals and category subtotals are live formulas, not pasted values.

general

Daily 7am briefing across Slack, Gmail, Calendar, Linear

Every weekday at 7am, scan Slack, Gmail, my Calendar, and Linear, and give me a priority briefing before my first meeting — what changed overnight and what needs me.

#SlackMGmail31Google CalendarLLinear

Scanned 6 Slack channelsovernight messages + mentions
Triaged 23 new emailsurgent vs. FYI
Read today's calendar5 meetings, 2 need prep
Checked Linear4 issues moved to your review
Scheduled the recurring runweekdays · 7:00 AM

Scheduled ✓ — I'll run this every weekday at 7:00 AM and drop the brief here. Today's headline: 3 things need you before 10am. Acme is waiting on your reply, the pricing doc needs sign-off, and your 9:30 needs 5 minutes of prep (notes attached).

Everything links to the source — one click to the Slack thread or email.
You can change the cadence or what it scans anytime in Scheduled Tasks.

general

Size the US SMB-analytics market and compare vendors

Size the US SMB analytics market, compare the top vendors by price and ICP, and build me a shareable briefing I can send to the team.

BQBigQuery

Researched 14 market sourcesreports, filings, pricing pages
Modeled TAM / SAM / SOMbottoms-up by segment
Compared 8 vendorsprice, ICP, and gaps
Compiled a sourced appendixevery number cited
Built market-sizing.htmlpublished, shareable

Here's the market-sizing briefing. TAM $18.4B, SAM $6.2B, three-year SOM $430M. I compared eight vendors — the clearest wedge is sub-50-seat teams that enterprise BI tools price out and overserve.

Every figure carries a citation in the sources appendix.
Published to a live URL — share the link, no export needed.

Projects

Workspaces the agent uses to group conversations, memory, and outputs.

general

Default project

3 tasks 2 artifacts

Q2-Reporting

Board metrics & dashboards

1 task 1 artifact

Revenue-Ops

GTM signal & feedback

1 task 1 artifact

Scheduled Tasks

The agent runs these on the cadence you set, then drops the results in the task.

1 scheduled task

Daily briefing

Every weekday at 7am, scan Slack, Gmail, my Calendar, and Linear, and give me a single priority brief.

general Active Next run in 3d

Live Artifacts

Documents, dashboards, and code the agent produces. Share to get a live URL.

campaign-report .html

Not shared

Marketing updated just now

Public

Q2-Reporting updated just now

Public

general updated 5d ago

feedback-themes .md

Public

Revenue-Ops updated 1d ago

daily-briefing .md

Public

general updated 6h ago

expenses .xlsx

Draft

general updated 2d ago

campaign-report .html Marketing Not shared updated just now

Q2-Reporting Public updated just now

general Public updated 5d ago

feedback-themes .md Revenue-Ops Public updated 1d ago

daily-briefing .md general Public updated 6h ago

expenses .xlsx general Draft updated 2d ago

Connect Apps and Data

Connect a source once. Secrets go into a vault, scoped per connection — the agent never sees a raw key.

Connected

Available — featured

Connect HubSpot

Opens HubSpot authorization in your browser — one click to wire.

Memories

Rules, lessons, and saved context the agent can reuse — yours to inspect and edit.

Global 1

Project · general 2

Select a memory file to inspect it.

lessons.md global

— Always reconcile Stripe MRR against the subscriptions table before reporting; they drift by ~2%.

— The boss prefers numbers first, then the narrative. Lead with the headline metric.

— Publish dashboards as live URLs by default — the team shares links, not screenshots.

lessons.md project · general

— Fiscal quarter starts in February, not January. Q2 = May–Jul.

— "North-star" metric is NRR; report it on every dashboard.

rules.md project · general

— Never write to production databases. Read-only credentials only.

— Currency is USD. Convert at the receipt date's rate.

Skills

Extend the agent's capabilities with task-specific skills.

What should we build?

No project

This computer

Ask first

Codex

Try a task

Code runs in the MindsHub Agents desktop app on your own computer, against your own repos and folders — this preview stops where the task would start. Get the desktop app

Add a dark mode toggle

Completed

Code · Desktop app Acme web app This computer

Add a dark mode toggle and remember the choice.

Plan

Add the toggle to Settings

Save the selected theme

Check light and dark mode

Agent activity Sample task · 2 changes

file change

Settings.tsx

Added an accessible Light / Dark toggle

file change

theme.ts

Remember the selected theme between visits

command

Run theme checks

Toggle, reload and keyboard checks passed

Added a dark mode toggle to Settings.

The app remembers your choice between visits.
Both themes keep text and controls readable.
Toggle, reload and keyboard checks passed.

Prepared sample result — no repository was changed.

The agent finished this turn. Review the changes or send a follow-up.

Workspace auto

Enter to send · ↑↓ history · Shift+Enter for a new line

Review PR #212 before I merge — rate limiting on checkout

Completed

Code · Desktop app Storefront API Work MacBook

Review acme/storefront-api PR #212 — rate limiting on the checkout endpoints. Correctness and security regressions only, not style. Run the suite and lint, report findings with file and line, and don't push or comment on GitHub.

Plan

Read the diff and PR description

Trace the rate-limit middleware

Run tests and lint on the branch

Report findings with file and line

Agent activity 14 actions · 0 changes · 6m

command

gh pr checkout 212

Switched to branch 'feat/checkout-rate-limit'

command

git diff main...HEAD --stat

6 files changed, 214 insertions(+), 31 deletions(-)

tool

Read src/middleware/rateLimit.ts

tool

Read src/routes/checkout.ts

command

pnpm test --filter checkout

Tests 48 passed (48) · 4.2s

command

pnpm lint

0 errors, 0 warnings

Reviewed PR #212 at e41c7b0. Two findings, no files changed:

[P1] Rate limit keys on req.ip — src/middleware/rateLimit.ts:42. Behind the load balancer every request arrives from the same address, so one busy tenant throttles checkout for everyone. Set app.set('trust proxy', 1) so req.ip resolves from X-Forwarded-For, or key on the session id.
[P2] Limiter state is per process — src/middleware/rateLimit.ts:18. The in-memory Map resets on deploy and is not shared across the three API replicas, so the effective limit is 3× the configured one. The Redis client in src/lib/redis.ts is already available.

Validation: pnpm test --filter checkout 48 passed; pnpm lint clean. Nothing pushed, no GitHub comments posted.

The agent finished this turn. Review the changes or send a follow-up.

Workspace auto

Enter to send · ↑↓ history · Shift+Enter for a new line

Fix checkout retry double charges

Completed

Code · Desktop app Storefront API Work MacBook

Tracing the checkout retry path…

I'll reproduce the timeout, follow the stored payment attempt, and check whether retries reuse the original charge.

Reading retry.ts + stripe.ts

const attempt = await createPayment(order);
// Retry creates a fresh attempt.

Tracing the retry flow…

Prepared demo · work in progress

Fix the double-charge bug when a checkout request is retried.

Plan

Reproduce the double charge in a test

Trace the retry path in src/orders

Add an idempotency key per order

Run tests, lint, open a draft PR

Agent activity 23 actions · 3 changes · 9m

tool

Read src/orders/retry.ts

tool

Read src/payments/stripe.ts

file change

src/payments/stripe.ts

+ idempotencyKey: `order_${order.id}_charge`

file change

src/orders/retry.ts

Reuse the stored attempt id instead of minting a new one

diff

test/orders/retry.test.ts

+ it('charges once when the first attempt timed out')

command

pnpm test && pnpm lint && pnpm typecheck

313 passed (313) · lint clean · tsc clean

Reuse the order’s idempotency key and stored payment attempt after a timeout. Added a regression test for the retry.

313 tests passing · lint clean · typecheck clean

The agent finished this turn. Review the changes or send a follow-up.

Finish with GitHub

Fix double charge on order retry after Stripe timeout

acme/storefront-api#211

Draft created Checks passing

Next request

Workspace auto

Enter to send · ↑↓ history · Shift+Enter for a new line

Add a CSV export to the orders list, same filters and auth

Completed

Code · Desktop app Storefront API Work MacBook

Add GET /orders/export.csv that streams the current filter set as CSV — same auth and filters as GET /orders, capped at 50,000 rows. Ask before adding a dependency. Test the header row and the cap, and open a draft PR once the full suite passes.

Plan

Read the orders route and filters

Add the streaming export handler

Cover the header row and cap

Run the suite, open a draft PR

Agent activity 19 actions · 4 changes · 12m

tool

Ask first · add csv-stringify? — declined

Reused fast-csv, already in package.json

file change

src/routes/orders.export.ts

New file · 71 lines

file change

src/routes/orders.ts

+ router.get('/export.csv', requireAuth, exportOrdersCsv)

file change

docs/openapi.yaml

+ /orders/export.csv — per the API change checklist

diff

test/routes/orders.export.test.ts

+ 2 tests: header row · 50,000-row cap

command

pnpm test && pnpm lint && pnpm typecheck

314 passed (314) · lint clean · tsc clean

Added the export as a streamed response, so a 50,000-row pull holds a few KB in memory instead of the whole result set. Draft PR #209 is up.

Changed src/routes/orders.export.ts — new handler; reuses buildOrdersQuery() so filters match GET /orders exactly, streams rows through fast-csv and stops at 50,000 with an X-Truncated: true header.
Changed src/routes/orders.ts:24 and docs/openapi.yaml — route registered behind the same requireAuth middleware and documented, per the API change checklist.
Changed test/routes/orders.export.test.ts — a header-row test and a cap test against a seeded 50,001-row table.

Validation: pnpm test 314 passed (2 new); pnpm lint and pnpm typecheck clean. You declined a new dependency; fast-csv was already installed, so nothing was added.

The agent finished this turn. Review the changes or send a follow-up.

Finish with GitHub

Add streamed CSV export for the orders list

acme/storefront-api#209

Draft created Checks passing

Workspace auto

Enter to send · ↑↓ history · Shift+Enter for a new line

Projects

Repositories, folders, skills, and defaults shared by coding tasks.

ProjectResourcesUpdated

Connectors

Connect developer tools once, then add them to any Code Project.

GitHub Start from issues, create draft pull requests, and follow reviews and checks.

1 connected

acme-dev Used by 1 project

Connected

Linear Start from Linear issues and post progress back to them.

Not connected

Skills

Shared engineering practice, versioned and ready for every coding agent.

MindsHub Engineering skills maintained by MindsHub 2

Acme engineering acme/engineering-skills · main 2

Growth KPI App

Acme Labs · Marketing · Synthetic sample campaign

Last Google Ads campaign

More leads each week, with brand search driving the most efficient conversions.

Spend$4,800Four-week campaign

Leads160104 qualified · 65%

Cost per lead$30$4,800 ÷ 160 leads

Leads grew every week

28 → 52

Weekly spend stayed at $1,200. Cost per lead fell from $42.86 to $23.08 as leads increased.

Brand search led efficiency

Cost / lead

Choose a channel to inspect leads and spend.

From impressions to qualified leads

2% click-to-lead conversion

Impressions240,000Campaign reach

Clicks8,0003.33% click-through rate

Leads1602% of clicks converted

Qualified leads10465% of leads qualified

Where the budget went

$4,800 total

Channel	Spend	Leads	Qualified	Cost / lead
Brand search	$1,200	80	64	$15
Non-brand search	$2,400	60	30	$40
Retargeting	$1,200	20	10	$60

What to do next

Protect brand search

It generated half of all leads using 25% of spend. Keep coverage steady while watching volume.

Improve non-brand intent

Test tighter keyword groups and landing-page copy before increasing the $2,400 allocation.

Refresh retargeting

Its $60 cost per lead was twice the campaign average. Test new creative and exclude recent converters.

Illustrative four-week dataset. Qualified leads are a subset of leads. Each conversion bar shows the share reaching that stage from the previous one.

POSTGRES + STRIPE + LINEAR · LIVE

Growth KPI App

Live · updating now

MRR

$128.4k

▲ 9.4% WoW

New signups (24h)

412

▲ 6.0%

Churn

2.1%

▲ 0.3 pt

Net revenue retention

112%

— flat

Signups per hour, last 48h — click a red dot to explore

65/hr

Requests per minute

842/min

Recent signups

RESEARCH · 14 SOURCES

US SMB Analytics — Market Sizing

Published · shareable

TAM$18.4BAll US SMB analytics spend

SAM$6.2BSelf-serve, < 200 seats

SOM$430MReachable in 3 years

Vendor comparison

Vendor	Entry price	ICP	Gap
Enterprise BI A	$70k/yr	2,000+ seats	Overserves SMB
Dashboard SaaS B	$24/mo	10–50 seats	No warehouse
Notebook tool C	$0 OSS	Analysts	Not self-serve
The wedge →	—	< 50 seats	Underserved

Sources: Gartner 2025, vendor 10-Ks, pricing pages, Census CBP. Every figure cited in the appendix.

Customer Feedback Themes — May

Prepared by MindsHub · sources: 42 Gong calls, 318 Zendesk tickets, 96 HubSpot notes

Ranked by attached ARR, not raw frequency. Each theme links to its source calls, tickets, and quotes.

1 · Onboarding friction$214k ARR · 28 mentions
2 · Pricing confusion$156k ARR · 19 mentions
3 · Missing Salesforce sync$98k ARR · 12 mentions
4 · Slow CSV export$41k ARR · 22 mentions

"We almost didn't make it past week one — setup took three calls with support." — VP Ops, $48k ARR account (Gong, May 14)

Recommended follow-ups

Ship a guided first-run; target time-to-first-value < 10 min.
Rework the pricing page around seats, not credits.
Prioritize the Salesforce connector — 5 deals waiting.

Daily Briefing — Friday

Generated 7:00 AM · Slack · Gmail · Calendar · Linear

Reply to Acme — they're waiting on the renewal terms.
#sales · 7:42 PM
Sign off on the pricing doc before the 11am review.
Gmail · Dana
Your 9:30 with Initech needs 5 min of prep.
Calendar

Meeting prep · 9:30 Initech

Renewal call. They asked about SSO last time — it shipped Tuesday. Up-sell angle: 15 seats added in May.

Open loops

4 Linear issues moved to your review overnight.
2 Slack threads in #product mention you.

Every line links back to the source — one click to the thread, email, or issue.

expenses.xlsx ⚑ 2 duplicates flagged

SummaryReceiptsCategories

	A · Date	B · Vendor	C · Category	D · Amount	E · Note
1	05-03	Linear	Software	$1,200.00
2	05-07	AWS	Infrastructure	$3,480.12
3	05-09	Delta	Travel	$612.40
4	05-12	Figma	Software	$144.00	⚑ possible duplicate of row 9
5	05-15	Notion	Software	$96.00
6	05-18	Uber	Travel	$58.20
7	05-21	Sweetgreen	Meals	$84.10
…		TOTAL		$24,318.00	=SUM(D2:D39)

created with MINDSHUB

One platform

One model catalog. Three ways to use it.

Use MindsHub’s full model catalog through Inference, Cowork, or Code.

Your control plane

MindsHub Console

Manage access, track usage, and set budgets in one place.

MindsHub Inference

Use MindsHub models in your favorite agents, or build your own apps.

Explore Inference

MindsHub Agents

Two modes

Cowork

Give tasks to our agent workspace. Research, analyze, and create.

Explore Cowork

Code

Build features, fix bugs, and review code with our Codex-based coding agent.

Explore Code

Create free account

Pricing

Start free. Pay for what you use.

An always-free tier and pay-as-you-go access to every model at provider rates.

Free

Always free, for knowledge work with agents.

$0 / month

No credit card required.

Create free account

The MindsHub Air model for non-coding work
Use in our agents or yours
Classify, route and score with Jev
Daily limits under fair use Free covers everyday work with agents, in MindsHub Agents or your own. For bulk automation or high-volume workloads, use pay as you go. Read the fair-use policy

Pay as you go

Every model, billed at provider rates.

Per use at provider rates

Prepaid balance. No expiration or minimum spend.

Top up

Everything in Free, plus:

Every model in the catalog
Caching savings passed on to you
No plan limits; pay per use
Auto top-up with a monthly cap
Usage and spend tracking

See full pricing and model rates →

Read the text version

What is MindsHub?

MindsHub gives you a choice of AI models for the software you build and the work your team does. MindsHub Inference connects your apps and existing agents to the model catalog through one API. MindsHub Agents is the agent workspace for research, analysis, content creation, and coding. Both share the same catalog, account, and balance.

How can you use MindsHub?

MindsHub Inference

Use MindsHub models in your favorite agents, or build your own apps.

Build AI into your own software or connect the agents you already use. Access models through OpenAI Chat Completions or Anthropic Messages with one MindsHub API key. Choose the model that fits each task, with paid usage billed at the published provider rates.

Explore Inference

MindsHub Agents: Cowork

Give tasks to our agent workspace. Research, analyze, and create.

Cowork uses Anton, the open-source agent harness behind MindsHub Agents. Choose a supported model, connect your apps, and turn their data into a brief or another artifact to share with selected teammates. App connections and sharing have their own access controls.

Explore Cowork

MindsHub Agents: Code

Build features, fix bugs, and review code with our Codex-based coding agent.

Code uses the open-source Codex coding agent. Choose a supported coding model from the catalog to build features and fix bugs in the desktop app. Cowork and Code are two modes of MindsHub Agents.

Explore Code

Can you embed MindsHub Agents in your software?

Talk to us about embedding MindsHub Agents into your product. Contact the team to discuss your use case, integration options, and commercial terms.

How do you manage access, usage, and budgets?

As more people and applications use AI, MindsHub Console gives your team one place to manage it. For model requests made through MindsHub Inference, including requests from MindsHub Agents using MindsHub models, the console brings permissions, model access, usage, and budgets together:

Roles: give people the permissions they need, such as creating API keys or viewing billing, without giving everyone full administrator access. A saved role is ready to assign; creating it does not assign it to a team.
Model access: manage access by provider or individual model, with rules for a team or the whole organization.
Usage: review daily spending by model, team, member, or API key, then inspect token usage and costs in the detailed breakdown.
Budgets: set budgets for teams, people, or API keys, choose the models and budget period, and track spending against the limit. Alert only warns at the limit while requests continue; Pause paid requests stops further paid requests at the limit.

These controls apply to model requests routed through MindsHub. They do not control bills from requests sent directly to a provider or grant access to repository files or CRM records.

How do you get started?

Create a free account, then connect your software through the API or use MindsHub Agents for coding and everyday work. The free tier includes the MindsHub Air model and Jev, TypeSafe’s decision model, with no trial period or credit card required. Jev is available through the Decisions API. The plans below explain included models, fair-use terms, and paid model rates.

Create free account

Free

Always free, for knowledge work with agents.

$0 / month. No credit card required.

The MindsHub Air model for non-coding work. Air uses a model selected for everyday knowledge work. The underlying model may change as better options become available.
Use in our agents or yours
Classify, route and score with Jev. Jev is TypeSafe’s decision model: it returns choices, scores and probabilities, not chat replies. Call it through the Decisions API as jev with your MindsHub API key.
Daily limits under fair use. Free covers everyday work with agents, in MindsHub Agents or your own. For bulk automation or high-volume workloads, use pay as you go. Read the fair-use policy

Create free account

Pay as you go

Every model, billed at provider rates.

Per use at provider rates. Prepaid balance. No expiration or minimum spend.

Everything in Free, plus:

Every model in the catalog
Caching savings passed on to you. When you resend content the provider already has cached, such as a long system prompt, it bills at the model’s lower cached-input rate, where the provider offers one.
No plan limits; pay per use
Auto top-up with a monthly cap
Usage and spend tracking

Top up

See full pricing and model rates

More information

Models Documentation Support Privacy Terms

Site index Full site text

Open this text as Markdown

All your teams’ AI spend. Under control.

See how MindsHub works.

Welcome to Acme Labs

Delegate a complete task

Use models in your code

Organization / Billing

Free allowance

Organization / Members

Organization / Roles & permissions

Organization / Teams

Organization / Workspaces

Organization / Model access

Organization / Audit log

API Keys

Traces

Let's knock something off your list

Projects

general

Q2-Reporting

Revenue-Ops

Scheduled Tasks

Live Artifacts

Connect Apps and Data

Connect HubSpot

HubSpot connected

Memories

lessons.md global

lessons.md project · general

rules.md project · general

Skills

What should we build?

Projects

Connectors

Skills

Try it with your team.

MindsHub Console

MindsHub Inference

MindsHub Agents

Cowork

Code

Start free. Pay for what you use.

Free

Pay as you go

What is MindsHub?

How can you use MindsHub?

MindsHub Inference

MindsHub Agents: Cowork

MindsHub Agents: Code

Can you embed MindsHub Agents in your software?

How do you manage access, usage, and budgets?

How do you get started?

Free

Pay as you go

More information