Plans

The plan is the structure of the work.

A live map of what you're building and why: goals, the capabilities that answer them, the technical contracts underneath, and the issues doing the work, all on one canvas.

A plan canvas: intents at the top, the capabilities that serve them, an impact metric and a commitment, the contracts underneath, and issue lists holding real tracker issues

Why Plans

Why it saves time.

  • 01

    The spec and the work can't disagree.

    The PRD is a view of the plan, not a copy of it, so there's no separate document to update.

  • 02

    Everyone sees the same picture.

    Product, design and engineering look at one canvas: goal → capability → contract → issues.

  • 03

    Progress is visible without a status meeting.

    Issue lists sit on the canvas and capability progress rolls up from them.

  • 04

    Decisions keep their reasons.

    New teammates, or you in six months, can read why something was built this way instead of guessing.

  • 05

    Mistakes are cheap.

    Every AI change is snapshotted, and a restore is one click that can itself be undone.

What it does

Inside Plans.

A plan canvas: an intent, the capabilities that answer it, their commitments, impact and contracts, and the issue lists doing the work

anedai AI

Let the AI do the typing.

In the plan's chat, the canvas agent creates and edits cards and links, records decisions, and traces where anything came from. Every change it makes is snapshotted, so a restore is one click.

  • Ask why it's built this way

    The answer comes from the decisions on the card, with the evidence behind them.

    “Why do we hide unauthorized actions instead of disabling them?”, answered: greyed-out controls kept getting read as bugs, with the capability linked
  • Expand in a sentence

    Ask for a contract or an issue list and it adds the cards and links under the capability.

    A SAML metadata contract and an Enterprise SSO issue list added under Enterprise SSO
  • Record the decision

    It records what was decided and why, on the card it shapes, and says so if it's already there.

    “Viewers get a read-only toolbar”, already recorded on Hide unauthorized actions, with its rationale

The plan, connected

One graph, from the goal to the issues.

Every atom links to the ones above and below it, so any issue can say which capability it builds and which goal that serves.

A plan's graph, one flow: the intent Admins see actions they can't use, its capability Hide unauthorized actions, and under it a commitment, an impact, a contract and an issue list of four issues

FAQ

Questions, answered.

Do I still need to write a PRD?

No. The PRD tab is generated from the plan's cards, so it can't drift from what's being built. The ERD tab does the same for the data model.

What if something is no longer relevant?

Use Mark obsolete instead of deleting it. It's dimmed and left out of rollups, and its history stays readable.

Can I undo the AI's changes?

Yes. The canvas is snapshotted whenever the AI changes it, and Restore brings back the canvas with its issues, sprints and milestones. The current state is saved first, so a restore can be undone too.

Where does a plan live?

Each plan belongs to a project and sits under a team, in Team → Plans, with Overview, Plan, PRD, ERD and History tabs.

One connected product memory for the people and AI building your product.

Bring your tools, your docs and your team. AnedAI keeps the story between them.