Cominty
enfr
01 — Agentic Portal

Your AI task-solving partner at work.

Built for operational and business teams. No setup, no code: connect your documents and tools, add your Skills, and create custom agents to execute and automate your professional workflows.

Pull the Q1 2026 sales report from @HubSpot and summarize performance by product
Try
Talk to SalesAsk for Demo

Trusted by teams running production AI agents on Cominty

  • Entrepreneur Invest
  • Groupe ADSN
  • Groupe Actual
  • OMS
  • Spinpart
  • Eight Advisory
  • Berger Levrault
  • Axys
  • Values Media
  • IAC Partners
  • Zcomme
02 — How you can use Cominty

Turn every request into a deliverable.

A reply is the easy part. Cominty produces the actual deliverable — ready to publish, present or plug into your stack.

app.cominty.ai/chat
Pull the Q1 2026 sales report from @HubSpot and summarize performance by product
Environment ready
Resolving context & permissions
Reasoning
Scoping the request
I’ll pull the Q1 2026 closed-won deals from HubSpot and break them down by product line.
Ran commandQuery HubSpot — closed-won deals, Q1 2026142ms
Sub-agent finished48s
Retrieve and normalise the Q1 2026 deal export
Cross-checking the pipeline
Read fileRead hubspot_q1_2026_raw.md — 78 deals47ms
Ran commandRecompute revenue per product line — subscriptions vs store62ms
12 bundles mix a subscription with gear. I’m splitting them pro-rata so nothing is counted twice.
Read memoryRead the Q4 2025 baseline for comparison11ms
Building the workbook
Loaded skillSpreadsheet — 3 sheets, one editable chart1.2s
Ran commandValidate formulas → 0 errors58ms

I'll pull the Q1 2026 sales data from HubSpot and break it down by product line.

Here it is — the Q1 workbook with 6 sheets, plus an 8-slide deck, Vertex-branded, with editable charts and speaker notes on each slide.

Vertex-Q1-2026-Sales-by-ProductExcel · 30 KB
Q1 2026 — revenue reviewby product
Subscriptions carry the quarter, gear softens€1.20MQ1 REVENUE+14%VS Q4 202578DEALS CLOSED
Appendix+5
Vertex-Q1-2026-Sales-ReviewPresentation · 8 slides · 1.1 MB

Two things worth flagging: subscription revenue is recognised on the billing date rather than on activation, and bundles that mix a subscription with gear were split pro-rata between the subscription and store lines.

Want me to add a churn and renewal view on the subscription lines, or break the store revenue down by sport?

HubSpot✕
1
Chat

Same pattern for every format: docs, sheets, decks, dashboards, sites, diagrams — generated in context, versioned, and shared with a link.

03 — Four ways to work

From the easy ask to the long, complex job.

It all starts with intention. Select the mode that fits the task, or hand over the wheel and let Cominty decide — call the right agent on the spot, launch a planner to map the work with you, or kick off Hive to run long, multi-cycle tasks autonomously.

Chat. The default for everyday questions, drafts and lookups. One agent, one pass, an answer in seconds.

05 — MCP connectors

Connect your daily work apps in one click.

Browse a growing catalog of ready-to-use connectors built on the Model Context Protocol, or add your own by URL — we detect the authentication automatically. Every connection respects your org’s existing permissions.

  • Gmail
  • Google Drive
  • Stripe
  • Jotform
  • Asana
  • ActiveCampaign
  • Twilio
  • Todoist
  • Xero
  • PandaDoc
  • Netlify
  • Smartsheet
  • Linear
  • SurveyMonkey
  • PayPal
  • Zoho Books
  • SignNow
  • Bitly
  • GoCardless
  • Razorpay
  • Process Street
  • Semrush
  • Mercury
  • Pendo
  • Workable
  • Pylon
  • ZoomInfo
  • Guru
  • Slack
  • HubSpot
  • Calendly
  • monday.com
  • ClickUp
  • MailerLite
  • Wix
  • Zoho CRM
  • Strava
  • Close
  • Intercom
  • Granola
  • tldv
  • Motion Creative Analytics
  • Spotify
  • Box
  • Customer.io
  • PagerDuty
  • Read AI
  • Base44
  • Grain
  • Cloudinary
  • Mixpanel
  • Miro
  • Zoho Projects
  • Metaview
  • Gamma
  • Databricks
  • Google Calendar
  • Notion
  • Airtable
  • Webflow
  • Shopify
  • Klaviyo
  • Fathom
  • Square
  • Quo
  • Docusign
  • Cognito Forms
  • Otter.ai
  • Attio
  • Apollo.io
  • Clay
  • Ticket Tailor
  • Krisp
  • Zoho Desk
  • NetSuite
  • Canva
  • Snowflake
  • PostHog
  • Circleback
  • Mem
  • DocuSeal
  • Dovetail
  • Egnyte
  • Outreach

352 live over MCP690 more on the way

  • 01

    1000+ connectors, one click

    CRM, data warehouses, ticketing, monitoring, project management, internal tools — ready to connect and to act on, not just to read.

  • 02

    Organization or personal scope

    Share a connection with the whole workspace or keep it to yourself; admins govern what gets shared.

  • 03

    Bring your own by URL

    Point at any remote MCP server. Secrets stay in the vault, never in a prompt.

Ask for a demo
06 — LLM router

Take control of your AI models.

Never be locked into a single vendor’s roadmap, pricing or downtime. Cominty puts every leading model behind one control layer, so you choose the best engine for each job — and switch the moment something better, cheaper or faster arrives.

ComintyLLM router — Finance Agent
Primary modelruns every request first
Claude Opus 4.8AnthropicAnswered · 1.4sMedium
Failoverstried in order when the primary degrades
1GPT-5.6 SolOpenAIMedium
2Gemini 3.6 ProGoogle DeepMindMax
3Mistral Large 3Mistral AIStandard
+ Add failover

The request goes to Claude Opus 4.8 first, as every request does.

Model catalog120+ models, every lab, one pickerSearch models…
  1. 0Claude Opus 4.8 (Medium)
  2. 1GPT-5.6 Sol (Medium)
  3. 2Gemini 3.6 Pro (Max)
  4. 3Mistral Large 3 (Standard)
  5. 4Kimi K3 (High)
  6. 5GLM 5.2 (High)
  7. 6Claude Fable 5 (XHigh)
  8. 7Grok 4.6 (High)
  9. 8DeepSeek V4 Flash (High)
  10. 9Qwen 3.5 Max (Max)
  11. 10Llama 5 405B (Standard)
  12. 11Claude Haiku 4.5 (Low)
12 / 120+Private endpoints appear in the same list
The labs behind the picker
  • OpenAI
  • Anthropic
  • Google DeepMind
  • Mistral AI
  • DeepSeek
  • Qwen
  • Moonshot AI
  • xAI
  • Meta
  • Z.ai
  • NVIDIA
  • + Your own endpoints
  • 01

    120+ native models

    OpenAI, Anthropic, Google, Mistral, DeepSeek, Qwen, Moonshot and more, ready to use on day one.

  • 02

    Bring your own model

    Connect private endpoints or custom weights over your own API — they appear in the same picker.

  • 03

    Always-on failover

    Primary model first, then an ordered fallback chain, so a provider’s bad hour is never your outage.

  • 04

    Cost control

    Transparency on cost per model and token usage, per agent and per member.

Ask for a demo
07 — Skills

Build a capability once.Use it anywhere.

A skill packages instructions, scripts and know-how into a reusable unit — publish it once and it becomes available to anyone with access, official or custom.

  • 01

    Build with Skill Creator

    Describe what you need in chat and let the agent draft, validate and version the skill for you.

  • 02

    Bring your own

    Upload a packaged .zip or import straight from a GitHub repo with a SKILL.md.

  • 03

    Shared org-wide

    Official and custom skills sit side by side, scoped by workspace or by organization.

ComintySkill lifecycle — one file, every chatReady in every chat
Ask for a demo
08 — Projects

Create a project.Share it securely.

Group conversations, skills, documents and integrations under one project — then hand it to a teammate, a client or a stakeholder without giving up control.

02Secured by defaultShared · org only

  1. Every chat, skill, document and connector attached to a project stays together, with its own history.

  2. Links respect your org’s permissions. Nothing is public unless you say so, explicitly.

  3. Set an expiry, restrict to invited members, or cut access entirely in one click.

Ask for demo
09 — Admin analytics

Drive your AI activity safely.

Admins get a live control tower over the whole org: who is using what, which tasks dominate usage, and exactly what each model costs — down to the token.

ComintyAdmin — usage & costs1d7d15dAll▦ Custom
65.5KTotal queries+12% vs last week
€4.2KTotal cost-6% vs last week
2.3BTotal tokens+9% vs last week
2,750Active users+140 this month
Queries over time
ChatHive
Cost per agent
  • Hive 35%€1,470.00
  • Chat 15%€630.00
  • HR Scoring Agent 12%€504.00
  • Finance Agent 10%€420.00
  • Operations Agent 9%€378.00
  • Legal Review Agent 7%€294.00
See more 9 other agents
Leaderboard
View full leaderboard →
#MemberWorkspaceShareQueriesCost
1AHAmira HaddadHR2.0%1,284€82.30
2MLMarc LefèvreFinance1.7%1,106€70.90
3ERElena RossiManager1.5%964€61.80

Every query, token and euro attributed to a member, an agent and a model. The figures above are one organization, one week.

Ask for a demo
10 — Create agents

Build a team of AI specialists.No code.

Custom agents capture a recurring job once — a role, its instructions, the models it runs on and the exact context it works from — so anyone on your team can trigger expert-level work on demand instead of re-explaining it every time.

ComintyAgents — create one, no codeLive
HR Scoring Agent is live
In the composer’s picker for everyone with access.
In every composer
HRHR Scoring AgentScore this week’s applications
Done · no code writtenAgent created
Built by the people who own the job

HR Scoring Agent is in the picker now. Pick it in any chat and it works from its own brief — five steps, no code, no engineer involved.

  • OP
  • SO
  • IT
Start free
  • HR Scoring Agent — 2 tools · 1 sources — Claude Opus 4.8
  • Finance Agent — 3 tools · 2 sources — Gemini 3.6 Pro
  • Operations Agent — 4 tools · 3 sources — GPT-5.6 Sol
  • Legal Review Agent — 2 tools · 4 sources — Claude Opus 4.8
  • Sales Ops Agent — 3 tools · 2 sources — Mistral Large 3
  • IT Support Agent — 5 tools · 2 sources — DeepSeek V4
11 — security by design, not by patch

Everything a security team asks for in the first call.

Cominty is built the way a CISO would build it: isolated by default, encrypted everywhere, infrastructure-agnostic, and never trained on your data.

Audited annually
AICPA · SOC 2 · TYPE II · AICPA · SOC 2 · TYPE II · SOC 2Type IISOC 2 Type II
ISO/IEC 27001 · 2022 · ISMS · ISO/IEC 27001 · 2022 · ISMS · ISO 270012022ISO 27001 2022
EU · GENERAL DATA PROTECTION · EU · GENERAL DATA PROTECTION · GDPRCompliantGDPR Compliant
EU · REGULATION 2024/1689 · EU · REGULATION 2024/1689 · EU AI ActCompliantEU AI Act Compliant
  • GDPR compliant — Data residency and processing controls native to every deployment. (DPA & sub-processor list)
  • SOC 2 Type II — Independently audited controls across security, availability and confidentiality. (SOC 2 Type II report)
  • ISO 27001:2022 — Information security management system aligned with the 2022 revision of the standard.
  • EU AI Act compliant — Commitments, measures and processes mapped to Regulation (EU) 2024/1689, with ethics by design. (AI Act compliance dossier)
  • Data & residency

    Encrypted

    At rest: server-side AES-256 on every stored artifact and secret. In transit: TLS 1.2+ enforced on every external connection.

  • Data & residency

    Never trains on your data

    No training, no fine-tuning — ours or our providers'. API-only, contractual no-training terms, zero data retention at OpenAI.

  • Data & residency

    EU-hosted by default

    AWS EU regions with data residency guarantees; every sub-processor listed in the Trust Center.

  • Isolation & operations

    Full audit trail

    Every agent action, tool call and approval is logged with who, what and when — exportable to your SIEM.

  • Isolation & operations

    Tenant isolation

    Strict per-organization data scoping, least-privilege tool access.

  • Isolation & operations

    Tested & monitored

    Third-party penetration tests at least annually, intrusion detection on the production network, MFA on every remote access.

Want the full documentation — pen tests, audit reports, controls status…? Everything a procurement or security review needs, in one place.

Go to Trust Center (opens in a new tab)
12 — Pricing

An offer built around how you actually use it.

No hidden tiers, no surprises — pricing scoped to your team’s usage from day one.

ComintyYour bill, in shape
  1. Platform feeOne line, predictable

    One predictable line for the portal, the harness, the connectors and the governance layer.

  2. LLM budgetMetered to real usage

    Matched to your real usage and steerable per agent, per team and per model.

  3. Per-seat licenceNone
  4. Hidden tiersNone
  5. TotalWhat you actually use
Two lines. Both of them yours to read, steer and cap.
What that buys
  • 100% cost transparency, down to the token
  • Scalable, monitorable and controllable spend
  • No per-seat quotes that balloon at scale

The platform fee covers the portal, the harness, the connectors and the governance layer. The LLM budget is the only line that moves — and you set how far.

Talk to an expert
13 — Frequently asked questions

What teams ask before their first day.

The five questions that come up in every first call, answered before you have to ask them.

A Q&A bot fetches an answer; Cominty acts. It plans multi-step work and takes action across your tools — updating a CRM record, creating a ticket, drafting a follow-up from a transcript — while respecting permissions and staying traceable. If it cannot do the task, it is search, not agentic.

Get started

Become an AI-driven enterprise today.

Connect your tools, add your skills, and start working immediately — no setup, no code.

Talk to SalesBook a call
  • 30mn demo on your own use case
  • Get a free trial
  • Talk with AI security experts