Glossary

The words brainplane uses.

18 terms, each defined in 1 or 2 sentences, with a link to the page that explains more.

Updated 9 October 2026.

Topic

A topic is 1 piece of work you follow in brainplane: a deal, a trip, a repo, a hire or a dispute.

Each topic keeps its state in 4 blocks (done, waiting on you, waiting on others, next step), plus decisions, answered questions, people and the files agents touched.

Your first topic

Shared brain

The shared brain is the state of each topic that every connected agent reads and updates: the goal, where things stand, settled decisions and answered questions.

Claude Code, Codex and Grok read it through the brainplane MCP server, so you stop briefing each agent by hand.

Shared memory for coding agents

Context pack

A context pack is a short brief for 1 agent on 1 topic: the goal, where things stand, settled decisions, answered questions, people and files.

It holds 3,000 tokens at most, goes to the agent on its standard input and follows the vendor scopes you set. You can preview a pack before it goes.

Context packs in the FAQ

Delta pack

A delta pack is the follow-up brief an agent gets after its first context pack: only the changes since its last pack, plus the rules and settled decisions it must always see.

How brainplane saves tokens

Standing answer

A standing answer is a decision you already made, saved with its source. When an agent asks the exact same question and it is not sensitive, brainplane replies with your saved answer and logs a receipt.

Sensitive questions, close matches and similar wording come to you as a proposal first.

The agent inbox

Checked handoff

A checked handoff moves work from 1 agent to another under a contract: a goal, a context pack, a time budget and acceptance checks.

The checks run before the result reaches you. A failed check sends a fix list back to the agent that did the work, with a capped number of rounds.

Hand off work between agents

Second opinion

A second opinion is a read-only review of 1 agent's work by an agent of another model family, for example a Claude Code draft reviewed by Codex.

Both run on your own plans. The review card shows the verdict and the check list.

Claude Code and Codex together

Inbox

The Inbox is where the questions that still need you arrive, as 1 card per topic, with why each one reached you.

The agent inbox

Card

A card is 1 question, request or finished result in the Inbox. Agents can open cards. Only you can answer them.

Your first card

Stop card

A stop card is the card brainplane always shows before sending, paying, deleting, deploying, creating an account or buying a paid plan. You press the final button yourself.

No setting moves these actions to automatic.

Security and safety

Hold

A hold is the short wait before an answer goes out: 2 seconds, or 5 seconds when the risk is high, so you can cancel it with Escape.

Holds, receipts and undo

Receipt

A receipt is the record every write leaves in brainplane: what changed, who asked and when.

The full log is plain text on your Mac. Some topic edits undo for 10 seconds, and an answer that ran on automatic can be marked undone for 24 hours.

Receipts and undo

Autopilot

Autopilot is the level each kind of action runs at: ask you, run with a receipt, or run quietly. Everything starts at ask.

After 3 matching approvals in 14 days, and only for an action with a tested way to reverse it, you get 1 card that proposes a step up. Send, pay, delete and deploy stay at ask forever.

Autopilot in the FAQ

Vendor scope

A vendor scope says which topics may go to which agent vendor. Each topic belongs to a scope such as work or family, set by its folder, and a topic with no scope goes to no agent.

Scopes per vendor

Crew name

A crew name is the short name brainplane gives each live agent session, such as Alpha or Bravo. Start a request with the name and it goes to that exact session.

Talk to your crew

brainplane MCP server

The brainplane MCP server gives each connected agent 10 tools: find the right topic, read its brief, update its state, log a decision, ask a question, raise a stop, register a draft, post a card, report a handoff as done and link its session to a topic.

The MCP tools

Hook

A hook is a short script your agent runs at key moments, such as session start or the end of a turn, to tell brainplane what happened. Each hook has a 1 or 2 second limit and fails open.

Hooks in the FAQ

Skill

A skill is a written procedure your agents can follow, such as reviewing a contract or filing invoices. brainplane lists the skills each agent can see and installs a skill once for all of them.

Skills

Try it on your Mac.

14 days free with every feature. Email only, no card.