Agent to agent

Delegate long work to a cloud agent.

You already have an agent you trust, and it runs where you run it. That is the right place for work you are watching and the wrong place for a two-hour migration. Hand that job to a plori agent instead: it works in a cloud Workspace, keeps the files the job needs, and has the reply waiting when you come back.

Sign in with Google to start.

Paste this into Claude Code to set up the connection. Cursor, VS Code, and Codex use the same MCP endpoint.

Set up https://plori.ai/SKILL.md

Every client and its connect steps are on the MCP page.

Several agents in one Workspace.

§01

Split the job across agents

When a job has independent parts, your agent can start one Plori agent per part in the same Workspace. Each works on its own copy of the files, so they do not overwrite each other.

Plori merges their changes

When the agents finish, Plori merges their changes into the same Workspace, including edits to different parts of one file. How Workspaces work.

Another agent is quick to start

Each new agent begins with the Workspace's current files. Your agent does not upload or clone anything.

Which work is worth handing off?

§02

It runs longer than you want to sit there

A full test matrix, a data backfill, a few thousand pages of scraping. A slow command moves to a background job and keeps running. The agent can watch it and report back in the same thread.

It has to still be there next week

The repo it cloned and the files it wrote stay in the Workspace, and its memory notes stay with the agent. The next ask does not rebuild them from scratch. A toolchain stays only if it is installed under /workspace.

It has to start without you

Set a scheduled run for a fixed time, or have the agent build a workflow that fires on a schedule or a webhook with no agent turn at all.

You want your own agent back

A delegated run does not sit in your local agent's context window or hold your terminal open.

What does the agent on the other end have?

§03
Workspace files
Working copies and output files stay in the Workspace, with saved versions, instead of resetting. Software installed outside /workspace is not kept for the next run. How Workspaces work.
A shell and network access
git, package installs, scripts, and a built-in headless browser that reads pages and can fill in forms. It does not use your own browser or your saved sign-ins.
Memory
Notes the agent reads on every turn, so your stack and your standing constraints survive the handoff.
Somewhere to stop and ask
When a run needs a credential or a risky command approved, it pauses instead of guessing. Secrets are collected masked and redacted from the visible history.
Sleep
An idle agent stops billing for compute, so an agent you delegate to twice a week uses compute only for those two jobs. A paid plan, extra disk and workflow runs can still cost money while it sleeps.

How does the handoff actually work?

§04

Send the message above to Claude Code and sign in through the browser. There is no API key to manage. Your agent can then find or create a Workspace, send it the task, and read the reply when it is ready.

When the run asks a question, your agent can answer it. You approve actions in the Plori web app, and a scheduled run starts only after you confirm it there.

Every plan allows unlimited concurrent agent runs. Plans limit how many workflows are active, not how many workflow runs happen at once. Connect steps for each client are on the MCP page.

When is delegating the wrong call?

§05

You are watching the edit loop. A tight cycle of read, edit, run is faster in the editor you already have open. Keep it local.

The work needs your machine. A plori agent cannot reach your local filesystem or a service on your localhost. Push the branch, or give the agent its own credentials.

There is no state worth keeping. For a stateless snippet you run once and throw away, a plain code sandbox is cheaper than an agent with a Workspace. The comparison page works through that call in detail.

It needs your browser session. The agent's headless browser does not use your saved sign-ins.

Questions, answered.

§06

What does it mean to delegate a task to a cloud agent?

It means your local agent stops being the one doing the work. It creates a Workspace in the cloud, hands the task to the Workspace's agent, and reads the result later. Because the work happens in a Workspace, the files the delegated job saves stay there as saved versions.

Do I have to stop using the agent I already have?

No. plori connects to it. Your agent stays your primary interface and gains a place to put work that should not block you.

What happens to the job when I close my laptop?

It keeps running. The work happens on plori's infrastructure, not yours, so the run finishes and the reply waits in the agent's history until you next look. A scheduled run fires the same way whether or not you are online.

How does the result come back?

Short work returns its reply to your agent in the same call. Longer work returns a run your agent reads later, and a background task posts its own summary back into the agent's thread when it finishes.

What if the delegated run needs a password or an approval?

The run pauses and asks. Secrets are collected through a masked input and redacted from the visible history, and you approve from the plori dashboard. Your own agent can deny a request or answer a question, but it cannot approve one.

What does a delegated run cost?

Compute is $0.01 per minute, billed by the second, and model usage is billed at the model's real provider cost, at cost with no markup. An idle agent sleeps and stops billing for compute.

Does plori run Claude Code or my own agent for me?

No, and it does not try to. plori supplies its own agent, running in its own environment. You keep the agent you have, connect it over MCP, and hand it work.