Plannotator v0.27.16

Plannotator is an open-source platform designed to review and validate plans and code generated by AI coding agents.

The platform provides a centralized interface where developers can:

  • Review AI-generated plans

  • Annotate proposed actions

  • Approve or reject tasks

  • Inspect code changes

  • Provide structured feedback

  • Improve agent outputs

  • Maintain human oversight

Rather than replacing developers, Plannotator aims to keep humans involved in critical decision-making during AI-assisted development.

As AI coding agents become increasingly capable, developers face a growing challenge: ensuring that AI-generated plans and code changes align with project requirements before they are executed. Plannotator addresses this problem by acting as a review layer between developers and AI coding agents.

Instead of generating code itself, Plannotator focuses on making AI-driven development more transparent and controllable. It allows teams to review, annotate, approve, reject, and refine agent-generated plans and code changes before they affect a project.

Download Plannotator v0.27.16 - Software Mirrors

Plannotator v0.27.16 for Windows

plannotator-win32-x64.exe | 151.41 MB

plannotator-win32-arm64.exe | 148.38 MB

plannotator-paste-win32-x64.exe | 93.23 MB

plannotator-paste-win32-arm64.exe | 90.2 MB

Plannotator v0.27.16 for macOS

plannotator-paste-darwin-x64 | 65.97 MB

plannotator-paste-darwin-arm64 | 60.51 MB

plannotator-darwin-x64 | 124.16 MB

plannotator-darwin-arm64 | 119.15 MB

Plannotator v0.27.16 for Linux

plannotator-paste-linux-x64 | 89.3 MB

plannotator-paste-linux-arm64 | 89.35 MB

plannotator-linux-x64 | 147.47 MB

plannotator-linux-arm64 | 147.48 MB

Plannotator v0.27.16 Source Code

Plannotator v0.27.16 Source code (zip)

Plannotator v0.27.16 Source code (tar.gz)

Plannotator v0.27.16 Release Notes:

Follow @plannotator on X for updates
Missed recent releases?
  • v0.27.15: Plannotator TUI and Herdr Annotate announcement, element context on pinpoints, HTML links open as linked documents, All files panel, Classic diff default
  • v0.27.14: Pi plan progress survives compaction, Codex threads across rollout files, WSL browser setting, Mod+E edit mode
  • v0.27.13: Open a review on a specific base (--base, --diff-type), symlink containment on /api/doc, CI flake fix, Amp decision relay
  • v0.27.12: Unified decision control, token hover cards, local-vs-remote diff, approval notes
  • v0.27.11: OpenCode server leak fix, durable local feedback archive, unknown-subcommand fix
  • v0.27.10: Auto-viewed files on scroll, annotation undo/redo, OpenCode 2 slash commands restored, npm 12 agent terminal fix
  • v0.27.9: WebMCP browser-agent tools, HTML refresh from disk, host seams, lazy renderers, Windows uninstall fix
  • v0.27.8: Pi keeps its prompt cache across plan transitions, thumbs-up returns to HTML annotation, embed picker seam
  • v0.27.7: Pi host crash fix on Windows, Call Flow tree cap, jj fork-point base, plannotator knowledge skill + llms.txt
  • v0.27.6: Live app annotation lands on Pi, one interaction model for HTML pages
  • v0.27.5: Annotate your running app, Agent TUI placement, collapsed lockfiles, VS Code theme fix
  • v0.27.4: Portable Guided Review exports, guides.show share links, guide CLI, jj Call Flow

What's New in v0.27.16

Diagrams are the theme of this release. Plans and documents with Mermaid or Graphviz fences now render in your color palette, on Mermaid 12, in a viewer you can zoom, pan, and comment on directly: a node, an edge, a sequence message, or the whole diagram. Twelve pull requests went in, three from the community, one from a first-time contributor. Code review learned to open a diff file without a repository, HTML annotation renders embedded sibling pages instead of a blank app, and a release-wide QA pass fixed a crash, a print regression, and a VS Code panel break before any of it shipped.

Themed diagrams

Mermaid diagrams used to render in one fixed dark-blue palette no matter which theme you chose. Now they follow the active palette and mode. Node fills come from your card color, text from your foreground, edges from your muted foreground, subgraphs from your muted surface, and the categorical fills that pie slices and git branches use are seeded from your palette's own accent colors. Every text-on-fill pair is checked against the 4.5:1 contrast rule and every line against 3:1, for all 78 palette and mode combinations that exist, so a diagram never becomes unreadable because a palette has a dark accent. Mermaid itself moved from 11 to 12.0.0. The visible change is layout: 12 uses the ELK engine by default, which routes edges orthogonally and packs subgraphs more tightly. Flowcharts, state, class, ER, and requirement diagrams re-lay out; sequence, gitgraph, and pie are unchanged. The runtime is larger, so the plan editor loads it lazily on the first diagram, and a plan with no diagram never runs it. Mermaid 12 also introduced a heavier default node shadow; this release tones it down and derives its color from your palette, light on dark themes and dark on light ones, with a Diagram shadow control in Settings → Display if you prefer none, or the original strength. (#1556, #1557, #1563)

Comment on any node, edge, or diagram

The diagram canvas is new. Every Mermaid and Graphviz fence renders inside a viewer with zoom, pan, fit, and keyboard controls, and a full-screen popout of the same viewer. Click a node, an edge, a subgraph, a sequence actor or message, a note, or a class relation, and the comment composer opens on that part. Click empty space and the comment attaches to the whole diagram. Each comment gets a numbered badge and a ring on its target, lists in the annotations panel beside your text comments, survives a reload and a theme change, and exports to the agent with its location: Diagram node Router (router), line 14. Interaction was shaped by hands-on use. Nothing highlights on a plain mouse-over, because hover targeting fought the pan hand; a click selects and a drag pans, with a small threshold so a shaky click still lands. Hold Cmd (Ctrl elsewhere) to preview the target under the pointer. Edges were nearly impossible to hit at their 1 px stroke, so every edge carries an invisible 14 px hit area, and the edge label box no longer swallows the click at the midpoint. Sequence diagrams, whose parts Mermaid gives no ids, got their own anchor family. A pinned comment resolves by id, then label, then source line, and shows an Unanchored chip only when the diagram no longer contains it. The viewer arrived from the commercial Workspaces app, where it was built first, and now ships in @plannotator/ui as the one diagram engine for both. (#1560, #1562)

Review a diff file, no repository required

plannotator review --patch-file change.diff opens the code review UI on a unified diff from anywhere: an email, a paste, a CI artifact, a remote agent's output. --patch-file - reads it from stdin. The server takes the patch as its snapshot and skips VCS detection entirely; staging, hunk expansion, base switching, and open-in-editor are hidden rather than left to fail, the header names the patch file, and a bad or empty patch says so. Reviews without the flag are byte-identical to before. @soundvibe wrote the feature as a first contribution, with the server degrading cleanly on every repo-dependent endpoint. The browser-side gating and the open-in fix were added on top before merge. (#1554)

Embedded HTML documents render

An annotated HTML page that embeds a sibling page, through <iframe src="prototype.html">, <embed>, <object>, or a src assigned by script at runtime, used to show a second Plannotator inside every frame. The annotated page has no URL of its own, so relative references resolved onto the Plannotator server and hit the app's catch-all. Now the served page carries a base URL pointing at the session's asset route, which covers static attributes, script-assigned ones, and relative fetch calls alike; the asset route serves sibling HTML with its query string intact; and a framed request for a missing file gets a small 404 page naming it, never the app. Embedded documents stay sandboxed with no access to the session API. An armed pinpoint click on an embed pins the frame itself. A release-QA check found that the framed 404 also fired for the app's own document when VS Code framed it, so the extension panel showed "Not found" on annotate sessions. Fixed before tagging: the 404 applies only to paths that name a file. (#1561, #1565)

References finds code in packages named vendor</h3> Code navigation excluded any directory named vendor, target, build, dist, or coverage at any depth, so a Java package like com.example.vendor.app was silently dropped from References. Names that can only be tool output stay excluded everywhere; the ambiguous ones are excluded only at the repository root, where they are build output, since ripgrep already honors .gitignore for nested copies. A follow-up scoped the exemption per directory so a search from that package never re-admits the root vendor/ folder. @buptwlh reported it with a minimal ripgrep reproduction that made the diagnosis immediate. (#1559, closing #1558, #1564)

Printing from a dark theme

Printing or saving to PDF from a dark palette put near-black diagram labels on near-black nodes, because the print stylesheet forces text dark for paper and Mermaid 12 renders labels as HTML. Printing now renders the light half of your palette for the whole page, diagrams included, and restores your mode afterward. Light-theme users see no change. (#1564)

Additional Changes

  • External annotation updates are validated. PATCH /api/external-annotations accepted any body; a diagramAnchor: null was stored and blanked the page. PATCH now runs the same field validators POST uses, on both runtimes, and the UI reads anchors defensively (#1564)
  • Element context reaches embedding hosts. The validator for pinpoint element context moved into @plannotator/core so a host can import it instead of copying it; nothing changes for Plannotator users. @FNDEVVE, closing #1521 (#1549)
  • Visual-explainer skill: a canonical diagram shell. Generated explainers now copy one zoomable diagram container with a clipping contract, so a zoomed diagram cannot paint over its caption. @FNDEVVE, addressing #1546 (#1551)
  • @plannotator/ui package publishes. 0.40.0 on core 0.25.3 carries Mermaid 12 and the theming; 0.41.0 on core 0.25.4 the diagram engine and the diagramAnchor field; 0.41.1 loads the engine lazily so a host's document read no longer ships CodeMirror and the viewer for a page with no diagram; 0.41.2 the shadow default. The HANDOFF names every export, the SVG id contract 11 to 12, and the one DOM-order change (edge paths now in declaration order). Consumers must add their own root overrides for lodash-es 4.18.1, since Mermaid 12's parser pins a version with two open CVEs and a package override does not travel

Install / Update

macOS / Linux:
bash
curl -fsSL https://plannotator.ai/install.sh | bash
Windows:
powershell
irm https://plannotator.ai/install.ps1 | iex
Claude Code Plugin: Run /plugin in Claude Code, find plannotator, and click "Update now". Pi: Update @plannotator/pi-extension to 0.27.16 and restart Pi. OpenCode: Clear cache and restart:
bash
rm -rf ~/.bun/install/cache/@plannotator

What's Changed

  • feat(ui): Mermaid diagrams follow the active color theme and mode by @backnotprop in #1556
  • feat(core): carry elementContext through html-anchor host helpers by @FNDEVVE in #1549
  • feat(ui): @plannotator/ui 0.40.0 with Mermaid 12.0.0 (ELK by default), loaded lazily by @backnotprop in #1557
  • fix(annotate): render embedded local HTML documents instead of loading the app inside every embed by @backnotprop in #1561
  • feat(ui): the Workspaces diagram viewer becomes the diagram engine (ui 0.41.0, core 0.25.4) by @backnotprop in #1560
  • fix(code-nav): stop excluding source packages named vendor/target/build by @backnotprop in #1559
  • fix(ui): load the diagram engine lazily, and keep the canvas controls out of the diagram; ui 0.41.1 by @backnotprop in #1562
  • docs(skills): canonical zoomable diagram shell for visual-explainer by @FNDEVVE in #1551
  • feat(review): add support for patch/diff files by @soundvibe in #1554
  • fix: three release-QA findings: PATCH validation, dark-theme printing, and the root vendor/ exclusion by @backnotprop in #1564
  • feat(ui): tone the Mermaid node shadow to 70 and derive its colour from the palette by @backnotprop in #1563
  • fix(annotate): scope the framed 404 to paths that name a file, so a VS Code session renders the app by @backnotprop in #1565

New Contributors

  • @soundvibe made their first contribution in #1554

Contributors

@soundvibe built patch-file review in #1554, a clean first contribution: the server refuses every repository-dependent endpoint with a clear error instead of crashing, nothing writes the patch to disk, and semantic diff works on the patch alone. The browser-side gating was layered on before merge, and the design underneath is his. @FNDEVVE landed two more, bringing the count to twelve: the element context validator move in #1549, which lets the Workspaces app share the same code instead of copying it, and the diagram shell reference for the visual-explainer skill in #1551, which fixes a class of caption overlap he had reported himself in #1546. The reports that shaped this release:
  • @buptwlh reported the vendor package exclusion in #1558 with a two-command ripgrep reproduction; the fix and its follow-up both use his exact directory shape as the regression test
Thank you. Plannotator gets better because you tell us where it falls short. Full Changelog: v0.27.15...v0.27.16

Key Features of Plannotator

Plan Review System

One of Plannotator's core capabilities is reviewing plans generated by AI agents before execution.

Developers can examine:

  • Proposed tasks

  • Implementation strategies

  • Agent reasoning

  • Planned file modifications

  • Workflow sequences

This visibility helps reduce unintended changes and costly mistakes.

Annotation Tools

The platform allows users to add comments, notes, and guidance directly to AI-generated plans.

These annotations can be used to:

  • Clarify requirements

  • Correct misunderstandings

  • Provide context

  • Guide future agent actions

AI Code Review

Plannotator extends the review process beyond planning by supporting inspection of generated code.

Developers can:

  • Review modifications

  • Analyze diffs

  • Leave comments

  • Request revisions

  • Validate implementation details

This workflow resembles modern pull-request review systems.

Human-in-the-Loop Workflows

A major design goal is ensuring that AI actions remain subject to human approval.

Organizations can establish review processes where important actions require validation before execution.

Open Source Foundation

Plannotator is open source, allowing teams to inspect, modify, and self-host the platform according to their needs.

This transparency is particularly valuable for organizations adopting AI-assisted software development.

User Experience

The interface is designed around review workflows rather than direct code generation.

Instead of interacting with a chatbot, users primarily:

  1. Receive agent-generated plans

  2. Review proposed actions

  3. Add feedback

  4. Approve or reject changes

  5. Monitor execution results

The workflow feels familiar to developers accustomed to pull requests, code reviews, and project planning tools.

Productivity Benefits

As AI coding tools become more autonomous, review processes become increasingly important.

Plannotator helps organizations:

  • Reduce risky AI actions

  • Improve code quality

  • Increase accountability

  • Preserve architectural consistency

  • Encourage collaboration between developers and AI agents

For teams adopting AI-driven development, these safeguards can be as valuable as the coding agents themselves.

Collaboration Features

The platform supports collaborative review workflows where multiple team members can participate in evaluating AI-generated outputs.

This allows:

  • Peer review

  • Team approval processes

  • Shared annotations

  • Collective decision-making

Such features are especially useful for larger engineering teams.

Performance

Because Plannotator focuses on workflow management and review rather than model inference, performance largely depends on the connected AI agents and integrations.

The platform itself is lightweight and primarily serves as an orchestration and review layer.

Open Source Advantages

Being open source provides several benefits:

  • Transparent development

  • Self-hosting capabilities

  • Custom integrations

  • Community contributions

  • Vendor independence

Organizations concerned about compliance, security, or proprietary workflows may find these advantages particularly appealing.

Limitations

Plannotator is designed as a companion tool rather than a complete AI development platform.

Common limitations include:

  • Requires external AI coding agents

  • Best suited for teams already using AI-assisted development

  • Smaller ecosystem than mature developer platforms

  • Additional review steps may slow rapid prototyping

  • Some users may prefer fully autonomous workflows

The software delivers the most value in environments where oversight and quality control are priorities.

Pros

  • Improves transparency of AI-generated plans

  • Supports structured review workflows

  • Human-in-the-loop design

  • Useful annotation system

  • Open source

  • Self-hosting support

  • Familiar review experience for developers

  • Helps reduce AI-generated mistakes

Cons

  • Not a standalone coding agent

  • Requires integration with AI development tools

  • Smaller community than established developer platforms

  • Adds review overhead to workflows

  • Best suited for teams rather than casual users

Who Should Use Plannotator?

Plannotator is ideal for:

  • Software development teams

  • Engineering managers

  • AI-assisted development workflows

  • Organizations adopting coding agents

  • Open-source projects

  • Teams prioritizing code quality and governance

It is particularly valuable for environments where AI-generated code requires oversight before reaching production systems.

Plannotator fills an increasingly important role in the AI development ecosystem by providing visibility and control over AI-generated plans and code changes. Its focus on human oversight, structured reviews, and collaborative workflows makes it a useful companion for modern coding agents. While it is not a replacement for AI coding tools themselves, it offers a practical solution for teams seeking greater confidence and accountability in AI-assisted software development.

Plannotator v0.27.16
Free
Software Informations:
Developer:

Operating System:
Windows / macOS / Linux
Date Added:
2026-09-19T12:28:23.583Z
Categories:

Post a Comment/Report Broken Link: