Jet Admin
Also known as: Jet Admin, JetAdmin, Jet, jetadmin.io, Jet Admin AI Agent Builder
No code app platform whose AI Agent Builder runs tool using agents over 200+ data sources and apps, inside workflows, app chat or Slack, Teams and email, with a template gallery, model choice and on premise deployment.
Jet Admin is a no code platform for admin panels, portals and internal tools that added an AI Agent Builder alongside its app generator. An agent is defined by instructions and a set of tools drawn from the customer's data sources and integrations, and it decides each next step itself and does not follow a fixed script. When a task needs structured execution, the agent can call a workflow, and a workflow can in turn contain an agent as a trigger or a step.
Agents live in three places: as a chat component inside an app, as a step in a workflow that starts on a button click, an API call, a schedule or a webhook, or in the Agents section of the workspace. Teams can also reach them in Slack, Teams or email by tagging @Jet. A gallery of ready made agent templates covers meeting prep, standups, support triage, lead enrichment, Snowflake and BigQuery analysis, CRM pipeline reports, inbox management and recruiting, each listing the tools it uses.
Governance comes from the app platform, with row and column level permissions, SSO and audit logs on the Business plan, and deployment on Jet Cloud, in the customer's own cloud account or fully on premise. Customers choose the AI model, and Enterprise can run self hosted models.
Confirmation before risky actions is handled in the agent's instructions, with no dedicated approval step. Testing follows a manual checklist that runs safe tasks, inspects tool calls and parameters, keeps write tools scoped to test records, and then revises and repeats.
Vendor details
Canonical URL
https://www.jetadmin.io
Category
Agent builder
Subcategory
AI agent builder on an internal tools platform
Funding status
Private and backed by Y Combinator. Round sizes are not established from first party sources.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Agents use data sources and integrations as tools: Postgres, MySQL, MongoDB, Snowflake, BigQuery, Supabase, Airtable, Google Sheets, Salesforce, HubSpot, Zendesk, Stripe, Gmail, Outlook, Slack, Jira, GitHub, Google Calendar, Firecrawl, Tavily and 200+ more, plus MCP servers and REST or GraphQL APIs. Models from OpenAI, Anthropic, Google Gemini and Groq are available, and Slack, Teams and email reach agents by tagging @Jet.
In practice
An operations team builds a data manager agent over Supabase that retrieves, creates and updates records from a chat component inside their admin app.
A sales team installs the CRM Pipeline Reporter template so a scheduled workflow produces a weekly Salesforce pipeline summary.
A support lead tags @Jet in Slack to have an agent pull account info, open tickets and recent activity for a customer.
Sources & related URLs
Related / legacy domains
Agentic Index coverage score
9.5 / 14 capabilities · 68%
| Integrations & Tool Calling | Full |
|---|---|
|
Agents take authenticated action through tools the builder connects, covering data sources (Airtable, Supabase, internal tables), integrations and external APIs, plus workflows. Jet Admin lists 200+ sources including Salesforce, HubSpot, Gmail, Slack, Stripe and MCP servers, and agents act through governed connections. Sourcedocs.jetadmin.io/ai-agents/overview/agents.mdread 2026-10-02 |
|
| Workflow Orchestration | Full |
|
Agents and workflows compose both ways. An agent can be created inside a workflow as a trigger or step, and agents can call workflows when structured execution is needed. The workflow builder offers branching, switch, if/else and iterator steps across APIs, databases and CRMs. There is no stated multi agent coordination. Sourcedocs.jetadmin.io/ai-agents/overview/agents.mdread 2026-10-02 |
|
| Knowledge Grounding & RAG | Partial |
|
Grounding comes through live tools, not a maintained retrieval layer. Agents query connected databases, APIs and CRMs, and the Internal Knowledge Agent template searches company docs and wikis through an MCP server with citations. There is no stated vendor maintained knowledge base, index or embeddings store for agents. Sourcejetadmin.io/build-custom-airead 2026-10-02 |
|
| Human Oversight & Guardrails | Partial |
|
Oversight runs through instructions and scoping, not a built in approval gate. In the example agent, the instructions say to delete records only after explicit user approval and to confirm before changes, the testing guidance is to keep write tools disabled or scoped to test records, and builders are told to scope permissions and review every action. There is no stated pause and approve step before an action commits. Sourcedocs.jetadmin.io/ai-agents/overview/agents.mdread 2026-10-02 |
|
| Security, Identity & Governance | Partial |
|
Access controls include row and column level permissions, SSO and audit logs (Business plan), with granular permissions on agents' connections. There is no stated SOC 2, ISO 27001 or comparable attestation, and Jet Admin says only that it is designed to meet complex compliance requirements. Sourcejetadmin.io/use-cases/ai-agents-automationread 2026-10-02 |
|
| Observability & Auditability | Partial |
|
Agent work is only partly recorded. Builders can inspect which tools an agent uses, the parameters it sends and the data it reads, and audit logs record user activity on Business. Testing can draw on audit logs and run history only when they are available, and there is no stated retained per run trace of agent actions. Sourcedocs.jetadmin.io/ai-agents/test-an-agent-safely.mdread 2026-10-02 |
|
| Memory & State Persistence | Partial |
|
Memory is conversation context only. Agents use instructions and past input to guide behavior, and there is no stated memory layer with a scope and lifetime across runs. Sourcedocs.jetadmin.io/ai-agents/overview/agents.mdread 2026-10-02 |
|
| Deployment & Data Residency | Full |
|
Customers can deploy on Jet Cloud, in their own cloud account or fully on premise, with self hosted AI models on Enterprise. There is no stated region selection on Jet Cloud. Sourcejetadmin.io/use-cases/ai-agents-automationread 2026-10-02 |
|
| Prebuilt Agents, Templates & Packs | Full |
|
A gallery of roughly forty customer adoptable agent templates, each with its own page and listed tools, includes Meeting Prep, Daily Standup, Customer Support, Lead Enrichment, Snowflake and BigQuery Data Analyst, CRM Pipeline Reporter, Inbox Manager, Recruiting Sourcer and PostgreSQL Query Agent. An Agent templates section in the documentation covers them too. Sourcejetadmin.io/build-custom-airead 2026-10-02 |
|
| Triggers & Channel Coverage | Full |
|
Agents can start from events and schedules, not only from a person in chat. Agents placed in workflows start on button clicks, API calls, scheduled jobs (every minute, hour, day or week) and webhooks. Users reach agents through an in app chat component and by tagging @Jet in Slack, Teams or email. Sourcejetadmin.io/build-custom-airead 2026-10-02 |
|
| Model Flexibility & Routing | Full |
|
Customers choose the model. The plans include a Choose your AI model option, Jet Admin promises every model out of the box with no vendor lock in, OpenAI, Anthropic, Google Gemini and Groq are connectable, and Enterprise can run self hosted AI models. Sourcejetadmin.io/pricing-copyread 2026-10-02 |
|
| APIs, SDKs & MCP Extensibility | Partial |
|
The Pro plan includes API access and two way GitHub sync, workflows can be started by API calls, and the open source Jet Bridge connects databases. There is no stated reference covering Jet Admin's own agents or platform objects, and MCP appears as a connection agents consume, not a server Jet Admin exposes. Sourcejetadmin.io/pricing-copyread 2026-10-02 |
|
| Testing, Debugging & Optimization | Partial |
|
Testing is a manual method, not a harness. The method is to run a normal request, then missing context, irrelevant and out of scope requests, inspect tool calls and parameters, record prompt, outcome and tool calls, and repeat after revising. Workflows have a native debugger. There are no stated stored, scored test cases or regression runs. Sourcedocs.jetadmin.io/ai-agents/test-an-agent-safely.mdread 2026-10-02 |
|
| Browser & Computer Use | Not documented |
|
There is no stated browser, desktop or remote computer control by agents. The nearest thing is Firecrawl and Tavily, available as tools for crawling and web search, which fetch pages and do not operate an interface. Sourcejetadmin.io/build-custom-airead 2026-10-02 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Product updates
When asked to build an app on an existing database or API, Jet Admin's AI App Builder now asks for access, inspects the real data structure and proposes a plan first. The user reviews and confirms that plan before anything is generated.
Bears on: Human approval / guardrails
View sourcePricing
Pro at $20 a month, or $16 billed yearly. A free plan is available.
A flat plan fee, with tiers sized by the AI credits each includes. There are no per user or per app fees.
Included quota
Credit allowances per plan are not stated. Free includes limited credits to build an app. All plans include unlimited apps, users and workflows.
What is public
Plan names, monthly and yearly prices, feature lists and a feature comparison table are published. Credit quantities and Enterprise pricing are not.
Billing mechanics
Flat monthly or yearly plan fee (20 percent off yearly), with AI credits included per plan rather than per user.
Cost watchouts
Plans are priced by the AI credits each needs, but how many credits a plan includes and what a credit buys are not stated. Credit rollover and on demand top ups are Enterprise features, so lower plans may hit a hard monthly ceiling. SSO and audit logs require Business.
Variable cost rationale
Jet Admin meters AI credits. Each plan includes an allowance, not stated in numbers, and building and agent work draw it down. Plan fees are flat and users are unlimited, so headcount does not move the bill. On Free, Pro and Business the allowance is the limit, since on demand top ups and credit rollover are Enterprise only and no overage rule is published. Heavier credit use means a higher plan.
Additional watchouts
Seat cost is not the constraint. Agent usage is, and the credit allowance behind it is not published.
Overage / add-ons
On demand credit top ups and rollover are listed for Enterprise only. No overage rule is published for Free, Pro or Business.
Sales call required
Mixed (some tiers require a call)
Free / trial
Free plan with limited AI credits and unlimited apps, users and workflows, plus a 14 day free trial
Lowest paid plan
Pro at $20 per month, or $16 per month billed yearly
Commercial notes
Free, Pro and Business are self serve. Enterprise is sold through sales, with an uptime SLA and a dedicated success manager.
Key ambiguities
Credit quantities per plan are unpublished.
Cancellation / refund
Turning off auto renewal keeps access through the end of the current billing period, with no cancellation fee.
Support SLA / resale
Free has email support, Pro email and Slack, and Business priority support with a 99.9% uptime SLA. Enterprise adds a dedicated success manager.
Missing data
Credits included per plan, credit unit cost, and the Enterprise floor.
Related vendors
- Lindy — Lindy is an AI teammate that lives in Slack, answers from your tools…
- Activepieces — Open source, MIT licensed automation platform where every one of 700…
- Agenta — Agenta is an open source workspace where teams build AI agents by…
- AgentX — No code platform to build, evaluate, and deploy multi agent…
- AI Library — No code platform for building and running AI agents, with a…
- Aigensei — Enterprise agentic workflow platform that runs business processes…
Alternatives to Jet Admin
The closest documented capability profiles to Jet Admin among agent builders tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Bika.ai10.0 / 14Fuller documented coverage on APIs, SDKs & MCP Extensibility
- Imbue9.0 / 14Fuller documented coverage on APIs, SDKs & MCP Extensibility
- Celigo10.5 / 14Fuller documented coverage on Security, Identity & Governance and Observability & Auditability
- MindPal9.5 / 14Fuller documented coverage on Knowledge Grounding & RAG and Human Oversight & Guardrails
- ToolJet10.5 / 14Fuller documented coverage on Security, Identity & Governance and Observability & Auditability
- AgentX12.0 / 14Fuller documented coverage on Knowledge Grounding & RAG and Human Oversight & Guardrails
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded