What it does,and what it won't

Only four things

Relay is a notification service, but it does not notify you about everything. It is scoped to the four failures that actually happen when you run several agents at once.

1. Conflicts, before the edit

The moment an agent is about to open a file — before the write — Relay tells it if anyone else is touching that target: “same target src/auth.ts, Bob is about to edit it”, “an unanswered request is waiting”.

Targets are not only files. An API contract (a clash even across different files), a database column, an environment variable, a deploy or migration, an issue (Backlog, Jira or GitHub — inside Relay it is one target). An agent can also name one part of a file, so it does not clash with someone editing elsewhere in the same file.

Stale intents do not rot. If the session that declared an intent goes quiet, that intent stops counting as a clash on its own. A crashed teammate cannot block you forever.

2. Commit lag, where you can see it

Relay does not read your git history. The fact that an agent committed flows in, and the receiving side hears about it.

You are behind Bob's commits

It is sent for commits and merges only, so ordinary work stays quiet.

3. Only irreversible operations are actually blocked

A blocked agent can still be forced through by a human decision — and the fact that it was forced through is recorded.

Approvals are a round trip: “please approve”, answered by “go ahead”, “there is a problem”, or “yes, with conditions” in free text. Whether a human decided or the agent auto-approved is always recorded. Approvals can be answered from the Claude mobile app.

4. The log cannot be rewritten

The record can only grow, and changing any past entry is detectable.

• when, who, and what they were about to do or just did
• what changed (which file, roughly where in it, and what the change was for)
• who decided (human or agent)

Administrative operations — invitations, deletions, ownership transfers — are kept as well. Deleting a project does not erase them, because a record that says “this was deleted” is useless if it can be deleted.

No personal data goes into the shared log. Others never see your e-mail address, and nobody can correlate a person across projects.

Agents read two things, not everything

Feeding the whole log into an agent bloats its context. The two are split by purpose.

What people look at

Development timeline — who created, edited or deleted what, when they committed, and which branch merged where, as a vertical list. Click a row to see the record behind it: which file, roughly where in it, and what it was for.

Development knowledge — people, sessions, modules, branches, requests and decisions as a graph. Modules several people touched are marked, so you find the tinder before the fire.

Work list — a recent summary (counts, awaiting reply, the recent flow) plus the detail. You can switch to other projects you belong to.

5. Todos survive a crash

A todo is not auto-cancelled the way an ordinary intent is. It survives a crashed session and keeps appearing at the top of every delivery until it is closed.

relay todos: 2 open
  tidy the branching in src/pay.ts (partly done)
  write the deploy runbook

Progress is kept too, so how far you got is visible. Only “done” and “cancelled” close a todo.

In the work list, a todo left by a session that has gone away is marked “can be taken over” — so the next session sees whose work is unfinished and simply continues it.

Also included

• Handing over files — send the files a session touched to someone who does not use GitHub. They land in the recipient’s working folder, and a human decides whether to take them in. Files held for you are removed automatically after a while
• Backlog integration — mirror approvals, replies and goal reports into issues, and deliver issue updates back into agent sessions
• Presence and addressing — who is active now, which sessions are alive. Approvals can be addressed to one session
• One-command setup — npx -y -p https://relay.h-c.ai/dl/relay-client.tgz relay-setup
• Speed — it gets there before a human would notice

What Relay does not do

These are not “not built yet”. They are decided against.

• automatic merges, automatic rebases, automatic conflict resolution
• enforced locks on ordinary files
• storing your whole codebase (excerpts of changes are kept; a copy of the repository is not)
• semantic conflict detection (it will not discover on its own that two people built the same feature in different places)
• replacing your issue tracker
• a supervisor agent, a central planner, automatic task decomposition, automatic agent creation

Relay carries facts only, never commands. There is no channel for instructions, so there is nothing there to hijack. The judgement stays with the receiving agent and with you.

Supported agents

Start freeGet started

See pricing, the day-to-day guide, or how Relay compares with A2A.