Emergent vs Replit Agent (2026)

The build agent wires outside services into each app in two ways. Integrations with a playbook follow a tested setup recipe for services such as Stripe, PayPal, Shopify, Razorpay, Paystack, SendGrid, Resend, Twilio, ElevenLabs and the OpenAI, Anthropic and Google models, and a few bill a small credit fee per run.

Anything else with a public API is coded in against the customer's own key. Teams can also register their own MCP servers under Manage Agents, giving the agent tools to read schemas, run queries, post to Slack or open Jira tickets while it builds. Servers are entered as JSON in the mcpServers format, and each tool list is cached for five minutes.

Sourcehelp.emergent.sh/connect-your-toolsread 2026-10-09

Connectors are integrations that Agent activates automatically based on the customer's prompts, and connector management across Core, Pro and Enterprise adds centralized admin controls, scoped access and audit tracking. Dedicated warehouse connectors cover BigQuery, Databricks, Snowflake and Microsoft Fabric. Use cases and build guides reach HubSpot, Slack, Gmail, calendar, Linear, GitHub, Ashby, Notion, Google Drive, Shopify, Stripe, Clerk and RevenueCat. External tools are also reachable through MCP servers listed in a directory, or custom MCP servers for anything unlisted.

Sourcedocs.replit.com chat/connectors, replitai/managing-connectors, connectors/warehouses, use-cases section, build/add-integrations and build/connect-via-mcp pagesread 2026-08-31

Every conversation runs as a long lived Temporal workflow in which the agent reasons, calls a tool, reads the result and repeats, up to 10,000 tool calls per message and four hours of wall clock time. A main agent plans the work and spawns subagents for testing, design, code review and integration setup, which run in parallel workflows and report back.

Three agents set the workflow when a job starts: E-1 builds step by step, E-2 checks that the pieces it builds fit together, and E-3 runs E-1 as a subagent under its own loop for large, open ended goals. Runs end on named stop reasons such as end_turn, handoff, max_iterations and insufficient_credits, and a handoff passes the task to another agent.

Sourcehelp.emergent.sh/how-the-agent-runs-workflow-stop-reasonsread 2026-10-09

Agent breaks work into tasks that can be reviewed, run in the background and applied back to the main version, and several features can be built at once through parallel tasks, with guidance on delegating safely and slicing work small.

Plan and Build modes are separate phases, Conversations and Projects are separate surfaces and a Conversation can be turned into a Project, and Routines schedule recurring runs that return to a thread. Projects hold multiple artifacts, and teammates can each start their own Agent thread on a shared project. Agents and automations built on Replit run on its cloud with Slack, Telegram and scheduled triggers.

Sourcedocs.replit.com task-system, build-in-parallel, plan-vs-build-mode, conversations-and-projects, projects-and-artifacts, invite-teammates and chat/routines pagesread 2026-08-31

Work reaches the builder when a person sends a message. A builder picks an app type (Full Stack App, Mobile App, Landing Page or Brainstorm), chooses an agent and types a prompt in the web workspace at app.emergent.sh or its mobile web app, and a new job can also start from an imported GitHub repository. In Plan mode the build waits until the plan is approved. On the self serve plans nothing starts a build on a schedule, a webhook or an event, so the agent works only while someone is driving it. The apps it makes can run scheduled jobs of their own once published.

Sourcehelp.emergent.sh/start-with-your-idearead 2026-10-09

Routines schedule recurring work from within a Conversation, using connected tools, creating projects or tasks in connected services and sending messages, with results returned to the same Conversation.

Routines run at intervals of one hour or more, with five active Routines per user on Core and ten on Pro, run only in Power or Max mode, carry a budget for each run that is checked after each step, and wait when due rather than queueing twice. Routines cannot schedule publishing. A Routines sidebar lists schedules, active status and run history with run-now and delete actions. Channels include Conversations, the prompt box, Voice Mode, the mobile app and Slack context.

Sourcedocs.replit.com chat/routines, home/routines, chat/conversations, chat/voice-mode and home/understand-your-workspace pagesread 2026-08-31

Each build draws on the job's own material: the whole conversation, the workspace code and files, the database structure, the key names in .env, and files attached to chat such as screenshots, Figma exports, PDFs, CSVs and specs. MCP servers add schemas, documents and live rows from outside systems. Pro raises the context window to one million tokens, long chats are compacted automatically, and a fork starts from an editable summary of the history. Context is gathered for each run from these sources, and the builder keeps no searchable index of a company's wider documents.

Sourcehelp.emergent.sh/a-tour-of-the-workspaceread 2026-10-09

Replit grounds Agent through per-request context and file uploads, Custom Instructions followed across Projects, Enterprise design systems held as a DESIGN.md, Agent Skills with supporting material, and project memory retrieved when relevant. Replit does not describe a searchable index it keeps over the customer's documents.

Design systems and Skills act as templates, Custom Instructions guide how Agent works, and the BigQuery, Databricks, Snowflake and Microsoft Fabric warehouse connectors supply data to the apps Agent builds. Grounding in the customer's own material comes from the context supplied with each request.

Sourcedocs.replit.com Memories page, 31 August read of chat/add-context, custom-instructions and connectors/warehouses, re-gradedread 2026-09-29

Each job keeps its chat, code and build state across sessions, so a builder can close the workspace and come back, and the agent reads the whole conversation on every turn. Workflow state persists in Temporal, so an interrupted run resumes where it left off, and a job paused at zero credits picks up again once the balance is topped up. Less carries between jobs.

A fork condenses the history into a summary the builder can edit, early details can fade after repeated compaction, and subagents start fresh on every turn. What the agent knows about a builder or team in a new job is whatever sits in the code, the forked summary or the files attached to chat.

Sourcehelp.emergent.sh/checkpoints-undo-anythingread 2026-10-09

Agent writes Memories while the user works and keeps them useful over time, with user memory scoped to one builder and workspace, project memory staying with one Project and retrieved when relevant, and custom memory following its own access and visibility settings. In Settings, Customization, Memory the user opts in, can turn Memory off, controls sharing with collaborators (off by default) and can view and update the Memory file, which is where memories are reviewed and cleared. Custom Instructions and Skills are kept separate from memory.

Sourcedocs.replit.com Memories pageread 2026-09-29

Plan mode, switched on in the composer, returns a plan card with two buttons, Request changes and Looks good, let's build it, and nothing is built until the builder approves.

Build mode, the default, starts as soon as a prompt is sent, Plan mode is unavailable in Brainstorm and E-3 sessions, and in autonomous runs a feature called Auto-HITL answers the agent's own questions with sensible defaults so the run never pauses.

Spending and release controls sit beside the plan step: a maximum credit cap per prompt that asks before going over, a budget per session, cancel from the chat menu, and a live app that changes only when someone publishes it again.

Sourcehelp.emergent.sh/talk-it-throughread 2026-10-09

Replit ships two gates in the run path. Agent proposes tasks with detailed plans that the user accepts or revises before they start, and each finished background task runs in an isolated copy and does not change the main version until the user reviews its work log, test results and live preview and chooses "Apply changes" or "Dismiss". Agent creates tasks automatically when it helps, and Replit does not say every edit in the main thread passes the same gate. Spend controls include usage limits, per user workspace spend limits and a budget for each Routine run.

Sourcedocs.replit.com Task system pageread 2026-09-29

Projects have four roles. Owners control billing, plan and deletion, Admins manage settings, members, the database and domains, Members chat with the agent and publish, and Viewers, on Enterprise only, see the app and its history.

Business adds single sign on through SAML or OpenID Connect and role based access control, Enterprise adds user groups and workspace wide two factor enforcement, and anyone can review and revoke active sessions. Emergent holds ISO 27001:2022 and SOC 2 Type I, with GDPR alongside them in its trust center, and its DPA sets out technical measures, breach notice within 72 hours and audit reports on written request. Secrets are encrypted at rest and data in transit, and customer personal data is kept out of model training under the DPA.

Sourcehelp.emergent.sh/team-roles-permissions-collaborationread 2026-10-09

Replit states SOC 2 Type II compliance and publishes an information security overview of its security practices, data protection and compliance standards.

Controls include SAML single sign on configured from workspace settings with an in product wizard that provisions the connection from pasted identity provider values, SCIM provisioning with synced group mapping to workspaces and roles, roles across Admin, Member, Guest and Viewer with custom groups on Enterprise only, separate account and workspace admin roles, audit logs with SIEM integration, Enterprise privacy and source control settings across the organization, per user workspace spend limits, and connector management with centralized admin controls, scoped access and audit tracking. A security checklist covers securing generated applications separately.

Sourcedocs.replit.com teams/information-security, teams/saml, teams/scim, teams/groups-and-permissions, teams/account-and-workspace-admins, teams/audit-logs, teams/enterprise-privacy-settings and replitai/managing-connectors pagesread 2026-08-31

While an app builds, the workspace shows the agent's activity live: each tool call with the arguments passed and the result returned, subagent logs nested under the main agent that expand to show their own calls, the step count for the turn and the credits being spent.

A checkpoint is saved after every agent turn, publish history keeps up to three earlier production images tagged with a timestamp, and Account Settings breaks credit use down by agent runs, publishes and Universal Key calls. Enterprise adds audit logs and credit usage reports with an analytics dashboard. How long the Enterprise audit log keeps entries is unclear.

Sourcehelp.emergent.sh/custom-agentsread 2026-10-09

Enterprise audit logs track and monitor security relevant actions across an organization, with SIEM integration. An Enterprise analytics dashboard reports activity, resource usage, published apps and costs. Connector management on Core, Pro and Enterprise provides centralized admin controls, scoped access and audit tracking. For each piece of work, the task system surfaces Agent's work as reviewable tasks that run in the background, Routines keep run history and status, and checkpoints record project state. Published apps are monitored separately after launch.

Sourcedocs.replit.com teams audit-logs, billing/teams-billing/analytics-dashboard, replitai/managing-connectors, task-system, chat/routines and build/publish-your-app pagesread 2026-08-31

Emergent runs as a hosted service, and customer data is processed and stored in the United States, the EU and India under its DPA. Builders choose the hosting region of a published app themselves, US or EU, in the Resources tab of Manage Publishing, with the EU priced higher. App databases run on MongoDB Atlas in the United States unless moved.

Enterprise deploys apps into the customer's own cloud through a VPC setup and supports a self hosted database, and any published app can move to the customer's own MongoDB by a manual runbook. The build platform and its subprocessors stay where they are whatever region an app uses, and EU only hosting for all data goes through a request to the privacy team.

Sourcehelp.emergent.sh/where-your-data-is-storedread 2026-10-09

Publishing offers autoscale, scheduled, static and reserved always on deployment types, with release settings, live verification and monitoring after launch. Mobile publishing runs end to end for iOS and Android, covering Apple Developer enrollment, App Store Connect upload and signing, TestFlight, Play Console upload and internal testing tracks.

Domains can be bought and connected in the product, and development URLs cover sharing work in progress. Enterprise offers private deployments, marketplace listings on GCP and Azure, and a Microsoft Azure partnership pairing Replit with Azure infrastructure. Replit does not describe a hosting region the customer can choose.

Sourcedocs.replit.com replit-deployments, build/publish-your-app, mobile publishing pages, build/domain-purchasing, development-urls and enterprise materialsread 2026-08-31

New jobs start from four app types, Full Stack App, Mobile App, Landing Page and Brainstorm, each on its own stack template (React with FastAPI and MongoDB, Next.js, Python only, or Expo for mobile), and from a choice of three build agents, E-1, E-2 and E-3, which are levels of one builder.

Integrations with a playbook package a tested setup recipe for each supported service, and the build page offers example prompts for websites, mobile apps, dashboards and SaaS tools. On Pro, builders make their own custom agents in a four step wizard covering the system prompt, tools and subagents. There is no gallery of finished apps or ready made agents to clone.

Sourcehelp.emergent.sh/how-apps-work-here-mental-modelread 2026-10-09

Enterprise custom templates standardize how an organization builds with Agent, alongside Enterprise design systems that configure Agent with the organization's component library, brand assets, design tokens and visual standards.

A quickstart covers building and publishing from a template, and Replit Design lets builders remix a featured template into a project and offers a library panel with starter templates, inspiration curated by Mobbin and reuse of the customer's own creations. Agent Skills package specialized knowledge that can be attached, installed in a project, or used to start one. Replit's public gallery shows templates with view and use counts.

Sourcedocs.replit.com teams/custom-templates, teams/custom-design-system, build/build-and-publish-from-template, design/start-from-a-template, design/library-panel and build/use-agent-skills pages, replit.com galleryread 2026-08-31

Two choices shape a build. The agent dropdown sets the workflow, and a separate model selector picks the LLM that powers it, offered again when a fork opens, so a builder can move to another model when a provider slows down. Pro adds Ultra thinking and Maxx mode, a deeper reasoning toggle that costs significantly more credits.

Model providers in the data path include Anthropic, OpenAI, Google Gemini, xAI and Groq. The Universal LLM Key, which reaches 44 or more models from seven providers on Emergent credits or the customer's own provider keys, serves the apps being built rather than the builder.

Sourcehelp.emergent.sh/understanding-models-e1-e2-e3-maxxread 2026-10-09

Intelligent model routing lets Replit pick an appropriate model for each task, with manual selection available when the customer wants that control. A model selector lets the customer choose an available model and set its Effort, though manual selection depends on the customer's mode and plan. In Replit Design the model behind the design surface can be switched to explore different directions, since each model brings its own sensibility and frontend strengths. Agent Modes across Free, Power and Max govern capability and spend, and Routines run only in Power or Max.

Sourcedocs.replit.com chat/auto-mode, chat/model-selector, chat/agent-modes and design/explore-with-different-models pagesread 2026-08-31

Emergent also works as an MCP server for clients such as Claude Desktop or Cursor, with example tools that create an app from a prompt, publish a workspace and fetch build or runtime logs, and personal access tokens with scoped permissions reach an Emergent API. The server's address and full tool list are not given alongside those examples, so how much of the platform outside code can drive is unclear. On paid plans the Universal LLM Key also works as a LiteLLM compatible proxy for calling models from scripts and CI, restricted by IP, and code leaves through Save to GitHub.

Sourcehelp.emergent.sh/mcps-connectorsread 2026-10-09

The Replit MCP Server at https://mcp.replit.com/server/mcp runs over Streamable HTTP with OAuth through protected-resource discovery, and its tools, including create_app_from_prompt, search_apps, resolve_app_by_name, list_apps and ask_question, create, find, inspect, update and publish Replit Apps from any MCP client. Native connections from ChatGPT, Claude and Slack run on the same server. Outbound MCP connections to external servers are a separate path, letting Agent use other tools.

Sourcedocs.replit.com Replit MCP Server and Connect through MCP pagesread 2026-09-29

Testing sits in the agent's own toolset. A testing subagent, called through a run_tests tool or when a builder asks for it, writes unit, integration and end to end tests for the app, runs them and reports coverage, and when a test fails the agent fixes the code and runs the tests again.

E-2 adds a workflow that proves the services it builds actually work together, a Codex review subagent lints and scans pull requests for security and style, and testing runs include visual regression scans.

Builders can ask for specific checks, such as the checkout flow end to end, results show in chat with what failed and why, and mobile apps get TestFlight and APK test builds on paid plans.

Sourcehelp.emergent.sh/debugging-testing-with-the-agentread 2026-10-09

Agent tests the application it built in a real browser, generates a report and fixes the issues it finds before handing the work back. That testing is Replit's own, separate from any test suite the customer already has. Further checks sit in the release path, with mobile testing on a real device through Expo Go and through TestFlight and Play Console internal testing tracks, a task system that lets work be reviewed before it is applied to the main version, and checkpoints that record state for rollback.

Sourcedocs.replit.com learn/build-with-agent, task-system, build/mobile-preview-testing and learn/security-checklist pagesread 2026-08-31

Agents can carry out tasks in a hosted web browser run by Browserbase, seeing and typing on pages, including ones the user is signed into, and content viewed or entered there is part of the data the platform processes. Firecrawl fetches web pages as a separate agent tool, and web search costs a small credit fee per query. Inside the app builder, the browser work Emergent describes is the preview server screenshotting the app under construction for the agent to check, so whether the builder drives the hosted browser on outside sites is unclear.

Sourceapp.emergent.sh/privacy-policyread 2026-10-09

A Browserbase connector runs managed headless browsers for fetch, search and full session automation, for cases where Agent needs to load a real web page, run JavaScript or interact with a site that blocks plain HTTP requests. Browserbase is a third party browser engine wired in through a connector, not a browser Replit ships. Agent's self testing also exercises the application it generated in a browser. Replit does not describe browser navigation, form filling or computer use of its own.

Sourcedocs.replit.com chat/connectors and replitai/managing-connectors pages, 31 August read, re-gradedread 2026-09-29