Rhei

Product context for coding agents

Product context for coding agents.

Rhei makes agents aware of your team's code, decisions, and prior work, then learns what matters as you ship.

$curl -fsS https://rhei.team/install | bash

Less guessing, fewer tokens, better edits. Local by default; shared memory is opt-in.

How it stays local

Copied. Paste into your terminal.

Read our vision and roadmap

01 / Flow

Most tools retrieve files Rhei follows the work.

Context, memory, traces, and impact carry forward across agents and sessions.

FIG 0.1

Context

The right files and prior work for the goal.

FIG 0.2

Memory

Past fixes, decisions, and dead ends, kept.

FIG 0.3

Trace

Sessions, edits, and checks stay attached.

FIG 0.4

Impact

Blast radius and proof gaps are visible early.

Less context waste.

Measured on replayed agent sessions, not synthetic benchmarks.

51.6%

fewer tokens

Right-sized context for review

89.6%

fewer tool calls

Less tool churn to reach signal

33.5%

faster runs

Same quality, less waiting

02 / Code Intelligence

Make agents aware.

Rhei gives coding agents task-specific context and memory across complex multi-repo work.

›

connected sources

cross-index

working context

selected for task

✓web/checkout/retry.ts
✓api/payments.ts
✓decision: retry boundary
✓prior fix: timeout edge
!proof gap: billing retry
·legacyRetry.ts

◆context ready for checkout retry

01

Selected working context

Rhei decides what belongs in the agent's working set for the task.

02

Reuse and cleanup

Similarity, accepted patterns, and stale paths are surfaced before new code is written.

03

Proof gaps

Missing validation and downstream impact stay visible while the agent plans.

03 / Agent Interface

Agents can use what Rhei knows.

Connect your repos, add Rhei once, and give your coding agents task-aware context in the tools you already use.

awareness
task
goalscopecurrent state
memory
prior fixdecisiondead end
Agents

◆agent awareness ready

04 / Agent Memory

It grows as you use it.

Goals, sessions, git history, and checks shape the next run. What worked returns. Dead ends fade.

memory trace

task: 

run 01

memory for next run

retry.ts
payments.ts
merged fix
billing check
rejected auth path

05 / Local

Local by default

Rhei starts on your machine. It maps your repo, builds local code intelligence, and exposes that context to coding agents over MCP. You can use local setup without an account. Pro features are optional. Hosted services for deeper context, team memory, and governance workflows only turn on after you opt in.

01

Local code map

Repo structure, symbols, relationships, and relevant context are built locally.

02

Works with your agents

Codex, Claude Code, OpenCode, and internal agents can use context where you already work.

03

Pro / Team is opt-in

Shared memory, hosted context, private deployment, and team controls require approval.

Read the security overviewData Use contractTalk to us about a private deployment

06 / Trust

Trust by architecture, not promises.

Teams that cannot leak code do not need another assurance page. They need a system where the safe path is the default path — and every exception leaves a receipt.

local mode

Source stays local

In local mode, code, the graph index, reports, and memory never leave the machine. No account required, no source upload — by architecture, not by policy.

never · metadata_only · selected_slices

Explicit upload modes

Cloud features only see what the configured mode allows. “never” means never — features degrade to local behavior instead of leaking source.

default

No training on your code

Rhei does not train models on customer code by default, and non-zero-retention provider paths are labeled before they can be enabled for a workspace.

audit trail

Receipts, not assurances

Cloud assist, context decisions, and memory promotions are receipt-backed. You can see what was sent, what was used, and why — per request.

enterprise

Runs inside your walls

Private Docker / VPC deployment with SSO, access and policy controls, and audit support. The same local-first engine, governed at org level.

workspace memory

Governed memory

Team memory is scoped per workspace and member. Knowledge is promoted by corroboration, decays when contradicted, and stays inspectable.

07 / Engineering blog

From the engineering blog

View all posts ->

Research

Reasoning is cheap. Awareness isn't.

Model tiers buy reasoning depth. The context layer buys sight. We measured what happens when a small model finally gets to see the whole change.

Rhei Team5 min read

Rhei

Start with awareness.

curl -fsS https://rhei.team/install | bashOpen

No account required. Pro trial is offered after local setup.