Skip to main content
Connect an existing JavaScript or TypeScript agent, run one real tool-using turn, and see exactly what Rippletide observed. Then add a rule in Observe mode and see what it would have blocked, without interrupting your agent.
Supported today: existing Node.js 18+ JavaScript/TypeScript agents and MCP servers. Python and other runtimes are not supported by this connection path yet.
This page is the standard CLI path for an agent repository. An MCP server follows the same starting point with the MCP guide.

​
1. Install and sign in

From the repository that contains the agent, install the CLI and sign in. Signing in is an operator step: it opens your browser and must be completed by a person with access to the Rippletide account.

​
2. Create the connection and get the repository-specific prompt

The CLI creates or links the agent, saves the project binding, writes the CLI connection configuration in a git-ignored .env, and gives you the exact setup prompt for this repository. It is safe to run again.
Use the generated prompt, not a generic installation prompt. It reflects the repository and tells the coding agent what to add. The CLI does not launch Codex, Claude Code, or Cursor for you.
For a headless coding-agent session, the human signs in first. The coding agent can then request the setup prompt from the repository root:

​
3. Apply the generated prompt

Paste the prompt from the CLI into Codex, Claude Code, Cursor, or your preferred coding agent. It installs rippletide-package, adds the repository-specific instrumentation, and declares the agent inventory. The prompt never contains your Connection key. Keep .env git-ignored and never replace a CLI-created .env with a block from another setup path. See Connect an agent for the app-based alternative and troubleshooting.

​
4. Run one real turn

Start your agent the way you always do — the wiring changes nothing about how it runs, and it still needs its own secrets (model keys, tool credentials) in place. Send it one message that makes it use a tool.

​
5. Verify

  • Open Harness management: you should see your agent’s identity, prompts, tools, and connections.
  • Open Runtime Events: the proof completes when the agent has registered, a heartbeat has arrived, and a tool-using turn has been recorded. (MCP servers run no agent loop, so a tool call is enough — no turn.)
From the repository, check both the configuration and the observed evidence:

​
6. Your first decision

This is the moment the product clicks:
  1. On the Rules page, pick the tool your agent just used and write a rule that would forbid what it just did.
  2. Save it and set the release to Observe mode.
  3. Send your agent the same message again.
The action goes through untouched — and the Rules page shows a would block decision with the rule that fired. You have just seen what enforcement would do, with zero risk to your agent. When you trust the rule, switch that release to Enforce: see observe vs enforce.

​
If something doesn’t appear

  • Node version: the SDK needs Node.js 18 or newer.
  • .env location: it must sit at the repository root (of the package that owns the agent, in a monorepo), and stay git-ignored.
  • Your agent’s own secrets: Rippletide adds none — model keys and tool credentials must be configured as before.
  • Is the process actually running? Evidence only flows while your instrumented agent is running.
  • Check the configuration from the repository root:
It reads the .env exactly like the SDK and checks the synced inventory. Use rippletide events --wait to wait for the complete runtime proof. If you’re stuck, report the command output to your Rippletide contact — never share the key itself.