How your work connects.
The Graph links conversations that share files, links, or commits, or that cover the same topics. Start from your projects, open their conversations, and follow the work to what it left behind: commits, failed tests, decisions, and memories.
Graphatlas
Follow what the work left behind.
The Work level puts one conversation in the middle, with the files it edited, the commits it made, and the decisions and memories it recorded. A test that failed in one conversation and passed later in another shows as one path.
Graph levelsEdited or relatedCommittedFailed or revertedPassed laterMemory evidenceDecided
Every link shows its evidence.
A conversation's connections are grouped by what links them. Open one to see the files, commits, and conversations behind it, and whether it was recorded or inferred from shared language.
Read connectionsSearch, then narrow it down.
Search matches the words in conversation titles. Filters narrow the graph by harness, model, kind of session, kind of connection, and dates, and List shows the same links in rows.
Filter the graphStill readable on battery.
On battery power, the Graph stops indexing and still answers from what it already has. Indexing picks up again once the computer can do background work.
When the graph pausesIndexing 128 of 412 conversations…Updates paused on battery power
- Levels
- 4
- Projects, Conversations, Work, and Local, one conversation's neighborhood.
- Kinds of work
- 6
- Conversations, files, commits, failures, decisions, and memories.
- Kinds of recorded link
- 12
- From shared files and commits to delegation and continuations.
- Around what you focus
- 2 hops
- Focus shows one or two hops around a project or conversation.
More of Hansard
Personal Sync
Keep your whole archive in storage you own, with every computer you use in step with the others.
Teams
Share whole conversations from the repositories you choose with your team, under rules every member can read.
Sharing
Send one conversation as a web page anyone can open, a PDF, Markdown, or a trajectory file.
Insights
Rank models on your own work with your own Pareto frontier, and see what you pay against list price and how the work turns out.
Running and Review
Run each project's agents from one place, with tabs, changes, and terminals, and answer everything that waits on you in one queue.
Reading a run
Follow any agent run step by step: a trace of its calls, the agents it started, its context and cost, and Create PR.
Slack bot
Start Claude Code and Codex runs on your Mac from Slack, approve each one in its thread, and get digests where you choose.
AI assistants
Let Claude.ai, ChatGPT, and other assistants search your agent conversations and memories, with your approval on your Mac.
Phone companion
Follow your agents from your phone, answer what they ask, and revive or start them while they run on your computer.
Accounts
Keep every Claude and Codex account in one place with its limits, route each conversation to one with allowance left, and cap API key spend.
Quality of life
Keep working when an account reaches its limit, and give every agent the MCP servers and skills you installed in one.
Put it on the record.
