Local-first knowledge graph

Shared context for your team and your agents.

Ingest a document once. Everyone on the team, and every AI agent they use, can traverse it, with the sources attached.

Get started See how it works

Free during the closed beta · runs on your machine · plain markdown you keep

Local checks for sensitive data. Your choice of AI provider. Privacy, explained

Person 7 Organization 3 Project 3 Meeting 4 Decision 4 Concept 4 Repository 2 Conversation 1 Source 4 Agent 1 Skill 1

Explore an entity

How it works

From a pile of files to a graph, in three steps.

01

Drop in what you already have

PDFs, decks, spreadsheets, email, Drive, Gmail, GitHub repos, exported chats. Screened for sensitive data before any model reads it.

PDF DOC XLS PPT MD

Plus any PDF, Office, or text file, a pasted link, or a local git repo. All sources ↓

02

An agent builds the graph, once

People, projects, decisions, meetings, and concepts each become one markdown file. Relationships are plain links. Every fact keeps its source.

wrote people/ada-patel.md
linked → projects/relay.md
updated people/README.md

03

Answers for people. Context for agents.

Teammates chat with the graph and get cited answers. Coding agents and assistants read the same folder over CLI or MCP. Nobody ingests the file twice.

✓ 3 claims · 3 verified

Connect a source, then start a sync when you want to bring in new changes. 7 connected · 3 on the roadmap · 13 file types

Shared context

Context is the layer AI-native companies run on. Yours lives in one place: a personal vault you can carry, or a team vault you share like a folder. What one person ingests, everyone can ask about, and every agent can traverse.

Each name in an answer is a node in the graph, with its sources one click away.

Hover the highlighted names →

What changed across Relay, and who owns the next steps?

            Read projects/relay.md
            Read people/ada-patel.md
            Read meetings/relay-release-readiness-review.md
          

Relay is ready for a focused partner review. Relay keeps the first rollout bounded. Ada Patel coordinates the outcome and Julian Morgan owns delivery.

The standing decision is Keep account context traceable; the next checkpoint is the release readiness review.

✓ 2 claims · 2 verified · $0.014

Context graph 14 of 35 files

Early results

Three pairs. One coding task. 20.6% lower run cost.

Three activated pairs on one privacy-detector coding task. Costs combine recorded and transcript-estimated usage. Quality was mixed: the blind judge preferred the graph arm in one of three pairs.

−20.6%

mean agent-run cost

$2.34 vs $2.94 across these three pairs

−33%

cache-read tokens

Mean reduction across the same three activated pairs

1 of 3

blind reviews preferred the graph

The control arms were more complete in two pairs

Explore a hypothetical 20% reduction

Comparable tasks per month 100

Baseline agent-run cost per task

Control-run range and mean from this one tested task.

Scenario baseline, per month

$294.00

Hypothetical reduction, per month

$58.80

Read the full analysisAssumes the same 20% reduction on every task. Three pairs on one task do not establish savings for other work. Excludes graph creation, maintenance, and human time.

Try it on your own questions

Stop asking around. Ask the graph.

Install openContext on your computer, add a source, and ask a question grounded in your own files.

Get started with the local beta

Answers cite the files behind them Runs on your machine, no telemetry Free during the closed beta Your teammates and agents read the same graph

Already installed?Install locally Read the evidence