agentleFS
waynesuttonwaynesutton/markdown-site/.claude/help/SKILL.md

help

Skill628 starsChanged 7 months agoMITView on GitHub

Reflection-first problem solving methodology for Convex development. Use before implementing any solution to ensure proper root cause analysis, 98% code confidence, and minimal change scope.

76 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use create-stories, story-done and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • lilangMaxSkill

    create-stories

    Break a single epic into implementable story files. Reads the epic, its GDD, governing ADRs, and control manifest. Each story embeds its GDD requirement TR-ID, ADR guidance, acceptance criteria, story type, and test evidence path. Run after /create-epics for each epic.

    Also in 9 projects that use this

  • lilangMaxSkill

    story-done

    End-of-story completion review. Reads the story file, verifies each acceptance criterion against the implementation, checks for GDD/ADR deviations, prompts code review, updates story status to Complete, and surfaces the next ready story from the sprint.

    Also in 9 projects that use this

  • +10 more from lilangMax

The file

Show its rules and text

We split this skill into the rules it gives an agent

22 rules. 0 are common, 19 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 57 edited versions we read keep these.

  • 19

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before implementing any solution, follow this process:

    line 12
  2. Distinctive:

    Proceed: Only move forward after clear understanding

    line 18
  3. Distinctive:

    Never assume: If anything is unclear, ask questions and clarify.

    line 20
  4. Distinctive:

    Use direct mutation calls with plain objects

    line 27
  5. Distinctive:

    Create dedicated mutation functions that map form fields to database fields

    line 28
  6. Very specific:

    Use indexed queries for ownership checks (not ctx.db.get())

    line 33
  7. Distinctive:

    Use timestamp-based ordering for new items

    line 35
  8. Very specific:

    Use Promise.all() for parallel independent operations to avoid write conflicts

    line 36
  9. Distinctive:

    Update Convex schema if needed

    line 61
  10. Distinctive:

    Only update files directly necessary to fix the original request

    line 62
  11. Distinctive:

    When tasks touch changelog.md, changelog page, or files.md:

    line 63
  12. Very specific:

    Run git log --date=short to check commit history

    line 64
  13. Distinctive:

    No placeholders or future months

    line 66
  14. Distinctive:

    Do not change UI, layout, design, or color styles unless specifically instructed

    line 70
  15. Distinctive:

    Never remove sections, features, or components unless directly requested

    line 72
  16. Distinctive:

    Always follow the site's existing design system

    line 79
  17. Distinctive:

    Never use browser default pop-ups

    line 80
  18. Distinctive:

    Use site design system components only

    line 81
  19. Distinctive:

    IMPORTANT: Do NOT create documentation files unless explicitly instructed.

    line 89
  20. Distinctive:

    Never use emojis in readme or app unless instructed

    line 100
  21. Distinctive:

    Don't write any code until you're very confident (98% or more) in what needs to be done.

    line 106
  22. Distinctive:

    If unclear, ask for more information.

    line 108

“Common” is measured: kept by at least 6 in 10 of the 57 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Core Development Guidelines Skill

Deep reflection and problem-solving methodology for full-stack Convex development.

1. Reflect deeply before acting

Before implementing any solution, follow this process:

  1. Reflect: Carefully consider why the current implementation may not be working
  2. Identify: What's missing, incomplete, or incorrect based on the request
  3. Theorize: Different possible sources of the problem or areas requiring updates
  4. Distill: Narrow down to 1-2 most probable root causes or solutions
  5. Proceed: Only move forward after clear understanding

Never assume. If anything is unclear, ask questions and clarify.

2. Convex implementation guidelines

Core principles

Direct mutation pattern:

  • Use direct mutation calls with plain objects
  • Create dedicated mutation functions that map form fields to database fields
  • Form field names should exactly match database field names when applicable

Best practices:

  • Patch directly without reading first
  • Use indexed queries for ownership checks (not ctx.db.get())
  • Make mutations idempotent with early returns
  • Use timestamp-based ordering for new items
  • Use Promise.all() for parallel independent operations to avoid write conflicts

Essential documentation

Functions:

Core concepts:

Authentication:

3. Change scope and restrictions

What to update

  • Update Convex schema if needed
  • Only update files directly necessary to fix the original request
  • When tasks touch changelog.md, changelog page, or files.md:
    • Run git log --date=short to check commit history
    • Set release dates to match real commit timeline
    • No placeholders or future months

What NOT to change

  • Do not change UI, layout, design, or color styles unless specifically instructed
  • Preserve all admin dashboard sections and frontend components unless explicitly told to update
  • Never remove sections, features, or components unless directly requested

4. UI/UX guidelines

Design system compliance

For pop-ups, alerts, modals, warnings, notifications, and confirmations:

  • Always follow the site's existing design system
  • Never use browser default pop-ups
  • Use site design system components only

Follow Vercel guidelines

https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/refs/heads/main/AGENTS.md

5. Documentation policy

IMPORTANT: Do NOT create documentation files unless explicitly instructed.

Banned unless requested:

  • README.md
  • CONTRIBUTING.md
  • SUMMARY.md
  • USAGE_GUIDELINES.md

You may include a brief summary in responses, but don't create separate documentation files.

Formatting rules:

  • Never use emojis in readme or app unless instructed

6. Code confidence requirement

98% confidence rule:

Don't write any code until you're very confident (98% or more) in what needs to be done.

If unclear, ask for more information.

Quick reference checklist

Before writing code:

  • Have I reflected on the root cause?
  • Do I understand what's actually broken?
  • Am I 98% confident in the solution?
  • Am I only changing files that need to change?
  • Am I preserving existing UI/features not mentioned?
  • Am I using the site's design system (not browser defaults)?
  • Am I following Convex mutation best practices?
  • Have I checked the relevant docs?

When uncertain:

  • Ask clarifying questions
  • Don't assume
  • Reference documentation
  • Narrow down to 1-2 most likely solutions

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.