Your business already has the answers — in Slack threads, GitHub pull requests, Linear tasks, your docs, and Salesforce service tickets. But those answers are scattered, hard to find, and often forgotten.

Eon brings them back to the surface.

Ask a question in Slack:

• "What did we ship last week?"
• "What's blocking the release?"
• "Summarize the latest GitHub pull requests."

Eon responds instantly, pulling from the tools you already use. No new UI, no new workflow — just answers in Slack, where your team already works.

• Unlock hidden value: Your Slack, GitHub, and Linear data already contain the insights you need. Eon makes them accessible.
• Faster decisions: Instead of searching or asking around, your team can get answers in seconds.
• Easy to run: The agent and MCP servers run statelessly in your own infrastructure, lightweight and simple to deploy.
• Seamless with Tiger: Eon uses a Tiger Postgres service (starting with a free plan). Prefer to self-host? Any Postgres instance with TimescaleDB works.
• Grows with you: Start with Slack and GitHub today. Add docs, Salesforce, and more tomorrow.

Available now

• Slack — Ask questions across your company's conversations and history
• GitHub (optional) — Summarize pull requests, issues, and activity
• Linear (optional) — Stay on top of project tasks and issues

Coming soon

• Docs — Search and summarize company knowledge
• Salesforce — Surface customer updates, opportunities, and account insights

Stay up-to-date with what's happening across your projects and team members with intelligent progress tracking.

Get comprehensive insights into recent developments, and cross-platform project updates.

Analyze technical changes, review code modifications, and understand the evolution of your projects.

• git
• Docker
• Accounts on:
  • Slack
  • GitHub (Optional)
  • Anthropic

git clone git@github.com:timescale/tiger-eon.git
cd tiger-eon
./setup

git clone git@github.com:timescale/tiger-eon.git
cd tiger-eon
cp .env.sample .env
cp mcp_config.json.sample mcp_config.json

1. Go to https://console.anthropic.com/settings/keys
2. Click "Create Key" and give it a workspace and name
3. Copy the value into .env where it says ANTHROPIC_API_KEY

1. Go to https://github.com/settings/tokens/new?description=Tiger%20Agent&scopes=repo,read:org
2. Change the token name, set expiration date, and then hit "Generate token"
3. Copy the value into .env where it says GITHUB_TOKEN
4. Set the GITHUB_ORG value in .env to your GitHub organization (e.g. if https://github.com/timescale was your org page, then set GITHUB_ORG=timescale).

1. Go to https://api.slack.com/apps/
2. Hit "Create New App" and select "From a manifest"
3. Select your org's workspace to install the org into
4. Copy https://raw.githubusercontent.com/timescale/tiger-slack/refs/heads/main/slack-app-manifest.json into the JSON input, changing the display_information.name and features.bot_user.name fields as needed
5. Click "Next" and then "Create" buttons
6. Scroll down on the page to "App-Level Tokens" and click "Generate Token and Scopes"
7. Give it a name and set the connections:write scope.
8. Copy the token into .env where it says SLACK_INGEST_APP_TOKEN
9. Click "Install App" on the left sidebar, and the click "Install to ..." button on this page
10. Grant the necessary permission and advance to the next page
11. Copy the "Bot User OAuth Token" (starts with xoxb-) into .env file where it says SLACK_INGEST_BOT_TOKEN

Follow the steps for seting up Slack Ingest/MCP, except:

• For step (4), use https://raw.githubusercontent.com/timescale/tiger-agents-for-work/refs/heads/main/slack-manifest.json instead, changing the same fields as needed for the name.
• For step (8), set the token to SLACK_AGENT_APP_TOKEN
• For step (11), set the token to SLACK_AGENT_BOT_TOKEN

1. Log into https://logfire.pydantic.dev/login (set region as necessary)
2. Under "Projects", click "New project"
3. Give it a name and hit "Create"
4. On next page, click "Generate Write Token"
5. Copy that into the .env file under LOGFIRE_TOKEN

To start configured services (based on COMPOSE_PROFILES in .env):

docker compose up -d

You can customize the MCP servers by editing a configuration file. To modify the system or user prompts, edit the Jinja2 templates.

To make the Slack MCP server more powerful, you can also restore a dump of your slack data, allowing eon to answer questions about historical threads or channels, and not just from when it was setup. To do so:

1. Follow the Slack Docs on getting an export of your workspace
2. Download the zip to your computer, and unzip it
3. Run ./import-slack.sh <path/to/slack/export> replacing the argument with the actual path to your dump.

Eon leverages the tiger-agents-for-work library, which provides Slack event handling with robust event queuing/retrying and TimescaleDB-powered event history.

graph TB
    %% slack app mention @eon
    U[Slack App Mention @agent]

    %% Main Orchestrator
    E[eon Agent]

    %% External Services (via MCP)
    GH[GitHub MCP]
    SL[Slack MCP]

    %% Flow
    U --> E

    E --> GH
    E --> SL

    %% Link to tiger-agent repository
    click H "https://github.com/timescale/tiger-agent"

    %% Links to MCP server repositories
    click GH "https://github.com/timescale/tiger-gh-mcp-server"
    click SL "https://github.com/timescale/tiger-slack"

• Agent + MCP servers — run statelessly in your infrastructure, easy to deploy and scale.
• Postgres — state. Use Tiger (with free plans) for seamless setup, or self-host Postgres if you prefer.
• MCP servers — connect to Slack history, GitHub.

Eon is open source under the Apache 2.0 license. Contributions welcome — file issues or PRs in GitHub Issues.