Era by Eon — Full simulated companies for your AI agents to run against | LaunchDaily
Era by Eon
Full simulated companies for your AI agents to run against
ai agent testing
synthetic data
simulation
mcp
benchmarking
developer tools
Era by Eon generates a complete simulated company — emulated SaaS systems (Salesforce, Slack, Zendesk, Jira, SharePoint, Google Drive, AWS, and more) populated with generated data, employees, and history — served over MCP servers and vendor-compatible REST APIs. Agents you already have can be pointed at it to develop, test, and benchmark against a full enterprise stack instead of mocks.
Key Features
One coherent company across every system
Each generated company is a single business projected into all its systems: the same customers, deals, and employees appear consistently across databases, object storage, Salesforce, Slack, SharePoint, Jira, and Google Drive.
Live systems over MCP and vendor-native APIs
Every system is a running instance, not a snapshot. Connectors serve both an MCP endpoint and the vendor's own REST API off one port, against one tenant's state — a write over MCP is visible when a REST client reads it back.
Realistic data distributions
Deal sizes, headcounts, and ticket volumes are drawn from published, real-world-shaped distributions (e.g. lognormal deal sizes with a cited median) rather than uniform noise, with a public scorecard of how much is estimate.
One identity per person across systems
Each employee is derived once — level, history, leave, and residue after departure — and appears as one identity across Slack, Jira, Zendesk, Salesforce, PagerDuty, and Asana, with consistent IDs behind every system.
Deliberate, declared mess
Optional opt-in defects — duplicates, placeholder values, inconsistent casing, junk rows — are dosed per column (capped at 30%), never touching IDs, keys, domains, or login emails, and every defect is declared in a manifest for benchmarking dedup, DLP, and access review.
Company history and incident narratives
Generated companies carry coherent timelines (first touch, security review, signed contract, onboarding, bugs, renewal risk) and shared incidents where multiple unrelated accounts file the same issue in their own words and timezones.
Console, CLI, and CI usage
Generate a company by answering five questions in the web console, via a three-command CLI (with plugins for Claude Code, Devin, Cursor, and Codex), or in CI through per-system Docker images on Docker Hub.
How It Works
1
Console generation
In the console, Era asks what the company sells, who it sells to, how big it is, whether it is living through a scenario, and which systems to project it into — then generates it while you watch. Nothing to install.
2
CLI and agent plugins
Install the CLI (curl -fsSL /install.sh | sh), add a plugin for Claude Code, Devin, Cursor, or Codex, then run 'era init' and 'era new --industry fintech --size mid --systems salesforce,zendesk,slack'. A skill file can be written for other clients via 'era skill --write'.
3
CI integration
Run systems in CI with per-system Docker images, e.g. 'docker run --rm -p 8115:8080 erabyeon/slack:latest'; images are published per system on Docker Hub and can be composed into an estate.
Pros & Cons
Pros
Systems are live and writable over both MCP and vendor-native REST APIs, not static fixtures
Deep cross-system consistency: one employee identity, one company graph, one coherent history across 11+ emulated systems
Data realism is engineered and published — timestamps follow working weeks and timezones, distributions cite their sources, and a public detector scorecard is provided
Opt-in, manifest-declared mess makes it useful for benchmarking dedup, DLP, and access-review agents rather than just toy demos
Runs in three modes: hosted console, local CLI with agent plugins, and Docker images for CI
Who It's For
Best for
AI engineering teams building and benchmarking agents that need to read and write realistic enterprise systems
Teams testing tools or automations across multiple SaaS systems (Salesforce, Slack, Zendesk, Jira, file stores) with consistent cross-system
CI pipelines that need a disposable, scripted company environment via Docker images
Not ideal for
Teams needing a staging environment that mirrors their real production data — Era's environments are entirely synthetic
Anyone looking for a static anonymized dataset rather than a live, running set of emulated systems
Use Cases
Developing and evaluating AI agents against a full, consistent enterprise stack instead of hand-written mocks
Benchmarking agents (deduplication, DLP, access review) against deliberately injected, declared data defects
Testing agentic workflows that span multiple systems — e.g. tracing a customer from Salesforce lead through Slack onboarding to a Jira bug
Pricing
The homepage states it is "Free for builders"; no detailed pricing plans or tiers are published.
Integrations
Salesforce
HubSpot
Zendesk
Gong
Jira
Confluence
Slack
SharePoint
Google Drive
Box
Dropbox
AWS
Deel
Eon
MCP servers
Claude Code
FAQ
What does Era actually do?
Era builds a simulated company — emulated SaaS systems, generated data and history, served over MCP and REST — so agents can be developed and evaluated against a whole stack instead of mocks.
Who is it for?
Teams that build, test, or benchmark agents against enterprise systems. The site targets builders with existing agents they want to point at a realistic, consistent multi-system environment.
How is it priced?
The homepage describes it as free for builders. No tiers or paid plans are published on the page.
How do I get started?
Either through the web console (answer five questions about the company and generate it while you watch, nothing to install), through the CLI ('era init' and 'era new' with agent plugins for Claude Code, Devin, Cursor, or Codex), or in CI via Docker images like 'erabyeon/slack:latest' on Docker Hub.
Built by Eon engineers who back up exabytes of enterprise cloud data, sizing generated cloud data to match real enterprise estates
Cons
Every environment is synthetic — it is no substitute for a staging system carrying real data
Runs on Eon's cloud, so generated companies are hosted by the vendor rather than fully self-hosted (only the CI Docker images run locally)
The emulated connector list is fixed and still expanding — additional connectors are listed as coming soon
Relatively early-stage product, launched recently with research partners still being announced
Which systems can be emulated?
The current list includes Salesforce, HubSpot, Zendesk, Gong, Jira, Confluence, Slack, SharePoint, Google Drive, Box, Dropbox, AWS, Deel, and Eon, with additional connectors listed as coming soon.
Can agents both read and write the data?
Yes. Each system is a running instance; agents connect over MCP and clients over the vendor's own REST API, and both operate against the same tenant state — a write over MCP is visible to a subsequent REST read.
Can I use it in CI?
Yes. CI images are published per system on Docker Hub and can be composed into a full estate, e.g. 'docker run --rm -p 8115:8080 erabyeon/slack:latest'.
Is it a replacement for a staging environment?
No. Every environment is synthetic, so it is not a substitute for a staging system that carries real data. It is intended for developing and evaluating agents against realistic emulated systems instead of mocks.
