> ## Documentation Index > Fetch the complete documentation index at: https://docs.pome.sh/llms.txt > Use this file to discover all available pages before exploring further. # Bring your own agent > Already have an agent project? Point Pome at it — one paste-prompt registers it and writes a pome.json, ready for its first scored run. The [quickstart](/getting-started) runs a bundled example so you can watch a scored run without any code of your own. This page is the next step up: you already have an agent in a repo, and you want Pome to test **that one**. One paste-prompt takes you from your repo to a registered agent with a written `pome.json` — the manifest Pome runs from — without you hand-authoring anything. The first scored run picks up from there. ## Start with the example Examples are the fastest way to learn the loop, so do the [quickstart](/getting-started) first if you haven't. Its paste-prompt takes the bundled `support-triage` examinee through a full run and shows you a real failing score (**33, not 0 or 100**) and the one-line fix. Five minutes, and you'll recognize every step the coach takes on your own agent next — this is a teaching step, not a detour. ## Before you start * The quickstart's setup, done once: the coach skills installed (`npx skills add pome-sh/digital-twins`) and the Pome MCP connected (`claude mcp add --transport http pome https://mcp.pome.sh/mcp`). * The **Pome CLI**, which writes and reads the manifest: ```bash theme={null} npm install -g @pome-sh/cli pome login ``` `pome login` opens `app.pome.sh`, signs you in, and stores a `pme_…` key in your macOS Keychain (or `~/.pome/credentials.json`). CI can set `POME_API_KEY` instead. * **Your agent's repo**, open in Claude Code — the one you want to test. It should talk to a service Pome has a twin for (GitHub, Slack, Stripe, Gmail, Linear), and you should know how it starts (e.g. `npm start`). ## Register your agent Open Claude Code **in your agent's repo** and paste: ```text theme={null} Connect this repo to Pome as my own agent — take it up to a registered agent with a written pome.json, and stop there. Do not run it yet. 1. Confirm I've already seen a scored run on the bundled support-triage example (the quickstart). If I haven't, tell me to do that first — the example is how the loop is taught. 2. Scaffold the manifest so I never hand-write it: run `pome init` in this repo (use `pome init --sdk claude` if this is a Claude Agent SDK project). That writes a starter pome.json. 3. Set `command` in pome.json to how THIS agent actually starts — detect it from package.json / the repo (e.g. `npm start`) and replace the scaffold default. Confirm the command with me before writing it. 4. Register the agent so runs scope to this project: `pome register agent "" --twins `. This POSTs the identity and writes the server-canonical agent.slug back into pome.json. 5. Show me the final pome.json (agent.slug + command + twins) and the dashboard line register printed. Stop here — the first run is the next step. Guardrails: keep every command plain text and show it to me before running; never write secrets or URLs into source. pome.json is safe to commit; the agt_ id register caches under .pome/ is not — leave it gitignored. ``` The coach runs the CLI on your behalf and shows you each command first. When it's done, your repo has a `pome.json` you never typed by hand and a registered agent on the dashboard. The registered `agt_…` id is cached in gitignored `.pome/link.json`, never in the committed manifest — `pome.json` carries only the portable `agent.slug`, so forking a repo can't collide two projects onto one cloud agent. ## What you have now * **`pome.json`** — `agent.slug`, the real start `command`, and the `twins` this agent may exercise. This is the manifest every run reads. * **A registered agent** on `app.pome.sh`, scoped to your team. That's the connection. Two steps remain before a score, and both are their own paths: 1. **Preflight the wiring.** Run `pome doctor` in the repo. It runs four checks in order — valid `pome.json` command, twin reachable, no production host surviving in agent source, egress locked to the twin — and stops at the first red with one named cause and one fix. A green preflight is the gate: don't launch a run until `pome doctor` exits clean. See [Troubleshooting](/docs/troubleshooting). 2. **Your first run.** With the wiring green, the coach suggests a first task matched to your agent and takes it to a scored report — the same run → read the failure → fix → re-run loop the example taught you. The [MCP workflow](/docs/mcp/workflow) is that loop, call by call. ## Next The example run this path builds on — see a scored run end to end first. The four wiring checks between a registered agent and a real run. CLI, skill, MCP, dashboard — the same loop from wherever you drive it. Where your registered agent, runs, and judge feedback live: app.pome.sh.