What problem it solves

AI coding agents produce plans, documents, messages, and code changes that need human review before the agents proceed. The README describes a workflow for marking up those artifacts, commenting on diffs, and returning structured feedback directly to the agent session.

Key features

• Annotate agent artifacts — Review plans, specifications, Markdown, HTML, messages, files, folders, and URLs with inline annotations. Rendered HTML can be annotated as-is, and agent replies can be reopened for review.

  Source: README

• Review code changes and pull requests — Inspect local diffs, remote pull requests, merge requests, patch files, and workspace changes in a browser. Reviewers can comment on diffs, suggest code, stage or unstage files, and return feedback to the agent session.

  Source: README, apps/review

• Connect feedback to agent workflows — Hooks and commands open the relevant review surface when an agent proposes a plan or completes work. Approval lets the agent proceed, while denial sends structured feedback that it can act on and revise.

  Source: README

• Manage multi-agent messages in the Inbox — The Inbox groups messages, questions, files, and guided reviews from several agents in one local window. Replies return to the session that asked without holding the agent session open.

  Source: README, apps/inbox

• Keep local workflows private by default — Plans, diffs, annotations, drafts, history, and configuration stay local by default, and the project does not collect usage telemetry or analytics. The README also documents the network boundaries for remote review, AI features, and sharing.

  Source: README

Who it may suit

Inference: This may suit developers who use AI coding agents and need to review their plans or code changes before handing decisions back to the agent. It may also suit teams that want an asynchronous way to share annotated artifacts and import feedback into agent sessions.

Topics to discuss

• Which review workflow should be prioritized first: plans, documents, local diffs, or remote pull requests?

• How should local annotations and feedback be shared when several agents or team members are involved?

• Which agent integrations and permission boundaries should the community prioritize supporting?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/backnotprop/plannotator#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.