PluginBench
Skill
Pass
Audit score 90

retro-marimo-pair

marimo-team/marimo-pair

Analyze pairing sessions to identify friction and improve marimo-pair skill or marimo._code_mode API.

What is retro-marimo-pair?

A conversational retrospective tool for marimo team members to review pairing sessions, surface friction points, and diagnose root causes in the marimo-pair skill or underlying notebook metaprogramming API. Use it after a session to turn frustrations into actionable improvements.

  • Guides structured session analysis to identify user frustration, inefficiency, errors, workarounds, and context loss
  • Walks through root-cause lenses (skill structure, gaps, misleading docs, API ergonomics, missing features, bugs, context window)
  • Produces clear diagnoses with contributing factors and trade-off considerations for the team to triage
  • Helps draft issues or discussions without auto-filing or committing to fixes
  • Supports optional skill edits or API changes only after explicit user approval

How to install retro-marimo-pair

npx skills add https://github.com/marimo-team/marimo-pair --skill retro-marimo-pair
Claude Code
Cursor
Windsurf
Cline

How to use retro-marimo-pair

  1. 1.Invoke the skill with a trigger like 'retro', 'what went wrong', 'improve the skill', 'session review', 'friction', or /retro-marimo-pair
  2. 2.Provide context about the pairing session (conversation history, specific complaints, or general review request)
  3. 3.Review the numbered friction-point summary and select which ones to dig into
  4. 4.Work through root-cause lenses for each selected point to converge on the most likely cause
  5. 5.Decide whether to draft the diagnosis as an issue/discussion for the team or (if explicitly requested) propose a skill or API change
  6. 6.Approve any proposed edits before they are applied

Use cases

Good for
  • Review a pairing session where Claude repeatedly misunderstood marimo-pair instructions and correct the skill docs
  • Identify that marimo._code_mode lacks a method needed for a common workflow and propose an API addition
  • Diagnose why a user had to work around a limitation instead of using the skill directly
  • Analyze multiple friction points from a session and prioritize which ones to file as team issues
  • Capture context-window problems where Claude forgot guard rails mid-conversation
Who it's for
  • marimo team members conducting internal retrospectives
  • Skill maintainers reviewing pairing session logs for improvement opportunities
  • Contributors diagnosing API ergonomics or documentation gaps

retro-marimo-pair FAQ

Should I use this to immediately fix the skill or API?

No. The default outcome is to diagnose friction and file an issue or discussion for the team to triage. Only if you explicitly ask will the skill draft a change, and even then it requires your sign-off before applying.

What if I find multiple friction points?

The skill surfaces all of them in Step 1, then lets you choose which ones to dig into. You don't have to analyze every point—focus on the ones that matter most to you.

Can I inspect the live marimo._code_mode API during a retro?

Yes. Run `import marimo._code_mode as cm` in your notebook scratchpad, then use `async with cm.get_context() as ctx:` to list and inspect all public methods and attributes.

What if the friction is in the marimo-pair skill docs vs. the underlying API?

The skill helps you diagnose both. It walks through lenses like skill structure, gaps, and misleading docs for the skill itself, and API ergonomics, missing features, and bugs for marimo._code_mode.

Will the skill edit files without permission?

No. The skill has a guard rail: it will never edit files in github://marimo-team/marimo-pair without explicit user approval.

Full instructions (SKILL.md)

Source of truth, from marimo-team/marimo-pair.


name: retro-marimo-pair description: >- Session retrospective for improving marimo-pair and marimo._code_mode. Use when the user wants to analyze friction from a pairing session, identify what went wrong, and brainstorm improvements to the skill docs or the underlying API. Trigger on: "retro", "what went wrong", "improve the skill", "session review", "friction", or /retro-marimo-pair.

Session Retrospective

You are helping a marimo team member review a pairing session to find friction and turn it into improvements. The target is always one or both of:

  1. The marimo-pair skill (github://marimo-team/marimo-pair)
  2. marimo._code_mode — the underlying notebook metaprogramming API

This is a conversation, not an automated report. You surface findings, the user steers which ones matter, and together you decide what to do about them.

Guard Rails

  • NEVER edit files in github://marimo-team/marimo-pair without explicit user approval.
  • ALWAYS start with session analysis (Step 1) — do not jump to solutions.
  • Present friction points before root causes — let the user choose which ones to dig into.
  • If the user invoked with a specific complaint, focus your analysis there but still scan for other friction in the background.

Step 1: Session Analysis

Review the current conversation and identify friction. Look for:

SignalWhat to look for
User frustrationCorrections ("no not that"), repeated attempts, backtracking, confusion, tone shifts
InefficiencyMultiple rounds for a one-step task, over-engineering, wrong API usage
ErrorsCompile-check failures, runtime errors, silent failures, wrong output
WorkaroundsUser or Claude working around a limitation instead of doing it directly
Context lossClaude forgetting instructions from earlier, re-asking things the skill covers

Present a numbered summary of friction points found. For each, note:

  • What happened (brief)
  • Where in the conversation it occurred (quote or paraphrase)
  • Initial category guess (skill structure / skill gap / API issue / etc.)

Then ask: "Which of these should we dig into? Or is there something I missed?"

Step 2: Root Cause Discussion

For each friction point the user selects, work through these lenses:

LensQuestionExample improvement
Skill structureWas the right info in the skill but hard to find? Buried in reference/ when it should be in SKILL.md?Promote to guard rail, restructure progressive disclosure
Skill gapWas information missing entirely from the skill?Add new section, example, or anti-pattern
Misleading docsDid the skill say something that led Claude astray?Correct the docs, add clarifying examples
API ergonomicsWas _code_mode clunky or unintuitive for this task?Propose API improvement (better defaults, clearer errors)
Missing APIIs there something _code_mode simply can't do that it should?Design a new API surface
API bugDid _code_mode behave incorrectly?Characterize the bug, propose fix or workaround
Context windowDid Claude forget instructions due to long context?Shorter, more prominent guard rails

Discuss each lens briefly, then converge on the most likely root cause with the user. It's okay to have multiple contributing causes.

Step 3: Diagnose & Capture

The goal of a retro is diagnosis, not a contribution. Based on the root cause, write up a clear diagnosis the team can act on — don't jump to proposing or authoring a fix.

For each friction point, produce:

  • Diagnosis — What went wrong and why it was frustrating, in plain terms
  • Contributing factors — Skill structure, gap, misleading docs, API ergonomics, missing API, API bug, context window (from Step 2)
  • Considerations — Trade-offs, open questions, or things that would need to be true for a fix to make sense. Note possible directions here, but frame them as considerations rather than committed solutions.

The default next step is to capture the diagnosis as an issue or discussion so the team can weigh it — not to immediately make a contribution. Concrete code changes (skill edits, API designs) come after an issue/discussion exists and the user explicitly chooses to go further.

Present the diagnosis and ask: "Want me to draft this as an issue or discussion?"

Step 4: File or Follow Up

Default: file an issue / discussion

Write it up clearly for the marimo team to triage:

  • Problem: What happened and why it's painful
  • Current behavior: What the skill or _code_mode does today
  • Considerations: Trade-offs and open questions (not a committed solution)
  • Example: A minimal snippet or quote from the session, if helpful

Leave the actual filing to the user — do not auto-file. This is the preferred outcome: surface friction for the team rather than ship a fix from the retro.

Only if the user explicitly wants to go further

A skill edit or API change should follow an issue/discussion, not replace it. If — and only if — the user explicitly asks to draft a change now:

  1. Read the target file in github://marimo-team/marimo-pair
  2. Show the proposed diff to the user
  3. Only apply after explicit sign-off
  4. After applying, verify SKILL.md stays under 500 lines (reference/ files have no limit)

Wrapping up

After completing the cycle for the selected friction points, ask if the user wants to revisit any remaining items from Step 1, or if the retro is done.

Key Files Reference

FilePurpose
github://marimo-team/marimo-pair/SKILL.mdMain skill instructions
github://marimo-team/marimo-pair/reference/execute-code.mdScratchpad & cell operation recipes
github://marimo-team/marimo-pair/reference/rich-representations.mdWidget & display patterns
github://marimo-team/marimo-pair/scripts/Bundled discovery & execution scripts

To inspect the live _code_mode API surface during a retro, the user can run in their notebook scratchpad:

import marimo._code_mode as cm

async with cm.get_context() as ctx:
    # List all public methods/attributes
    print([x for x in dir(ctx) if not x.startswith('_')])
    help(ctx)