Skip to content
Get packs

Plan Canvas Review

Lets the agent open plans in a local browser review page where you annotate, chat, and approve or request changes

Source Download skill
Repository
affaan-m/ECC
Stars
295,127
Licence
MIT
Files
2 files

What it does

When installed, the agent can open a plan or HTML/Markdown document in a local browser page on your computer. You can click or highlight specific parts, leave comments, chat, and press Approve or Request changes. The agent waits for your feedback as structured data and revises the document in place, instead of you typing long instructions in the terminal. It also supports Mermaid diagrams in plans.

Great for

  • Developers who use the /plan workflow and want a visual approve-or-revise step before implementation
  • Anyone reviewing a design, report, or plan where pointing at a specific element is easier than describing it
  • Teams using Claude Code, Codex, or similar agents with ECC installed

Skip it if

  • Code review of diffs or pull requests, which the skill explicitly excludes (use a code-review tool instead)
  • Reviewing live or remote web apps and URLs, since it only serves local artifact files; not useful for this
  • Users who expect the agent to proceed without a running shell: it depends on background commands and a local server on 127.0.0.1:4517, so it may fail in sandboxed or restricted environments

Try asking

  • “Open my plan in the browser so I can annotate it and approve or request changes”
  • “Present this HTML mockup for review and let me point at the parts to change”
  • “Show the architecture diagram in the plan and wait for my feedback before implementing”
  • “Open the report in the browser; I'll leave comments and chat with you there”
Step-by-step guide: how to use Plan Canvas ReviewInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 2 original files, the MIT licence and an attribution note.

    plan-canvas.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip plan-canvas.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip plan-canvas.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Ask the agent to run the await command in the background so feedback is not missed when the turn ends
  • Keep the artifact and any images or CSS next to each other and reference them by relative path
  • Tell the agent to end the review when you are done so it stops listening

What's inside

  • agents/openai.yaml
  • SKILL.md

Redistributed unmodified from affaan-m/ECC under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.