lnr-plan
Claude Code slash command that turns a Linear issue or free text into one reviewable HTML plan with risks, open questions and a diagram
Overview
lnr-plan is a Claude Code custom slash command, /lnr-plan, that turns a Linear issue, another tracker item or a free-text request into a single self-contained HTML plan. You never write HTML yourself: a fixed template engine renders the page, and the command file tells the agent how to read your issue and your code and fill in the plan-data blocks. The plan has an executive summary, a security and risks tab, an open-questions tab where reviewers pick suggested answers inline, a domain-by-domain impact breakdown (UI, API, modules, services, models and schemas, cloud), a Mermaid chain diagram and a printable full plan. Reviewers highlight text to leave comments. It is a very new project by Eyal Ronel, with a couple of commits and no releases as of November 2026.
The Verdict
Who Should Use lnr-plan?
Best For
- Teams on Linear and Claude Code who want a plan reviewed before any code is written
- Sharing a readable, printable plan with product managers and other non-developers
- Codebases with clear business domains that are worth mapping once
Not Ideal For
- Companies that need a clear open-source license before adopting code
- Anyone wanting a mature, supported tool with releases
- Teams who want a full spec-to-tasks workflow rather than one plan page
What's Great
- No build step; the output is one HTML file you can open or send
- Broad plan structure: risks, open questions and impact per domain
- Highlight-to-comment review flow for whoever reads the plan
- Works from free text too, so a Linear issue is optional
Watch Out For
- No license file, so reuse rights are unclear
- Very early: two commits on main and no releases
- Ships placeholder domain data; you must run the setup prompt and check its first draft
- Written against Linear tool names; other trackers need an edit to the command
Pricing
View all features & details
How It Works
- Copy three files (command, HTML template, world-map JSON) into
.claude/commands/ - Run the setup prompt once: the agent reads your repo and lists its real domains and modules
- The world map then stays fixed and is edited by hand, not regenerated per plan
- Each run fills the template's plan-data blocks from the issue and the code
Optional Extras
- Linear issue linking (issue badge shown only when linked)
- A "Send notes to Conductor" button and deep links
- A browser-automation tool for the final verification pass
How It Compares
| Feature | lnr-plan | GitHub Spec Kit | OpenSpec |
|---|---|---|---|
| Output | One reviewable HTML plan | Spec, plan and task files | Change proposals and specs |
| Tracker input | Linear or free text | Free text | Free text |
| Review flow | Highlight-to-comment | Pull request review | Pull request review |
| Maturity | Two commits | Established | Established |
Sources: lnr-plan README · SETUP.md · see also GitHub Spec Kit, OpenSpec, BMAD Method and Claude Code
Is this your tool? Add a badge