agentleFS
CodePhobiiaCodePhobiia/claude-roblox-game-studio/.claude/skills/doctor/SKILL.md

doctor

Skill9 starsChanged 7 months agoMITView on GitHub

Run a comprehensive health check on the repo. Verifies README counts match reality, promised files exist, no unresolved placeholders, wiki index is synced, toolchain configs are valid, and agent roster is accurate. Use after making structural changes or before publishing.

86 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use uninstall, cancel 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.
  • AnthropicSkill

    uninstall

    Uninstall a community skill that was installed via the hub. Confirms before deleting files, refuses to touch first-party plugin skills, and logs every action. Use when the user wants to fully remove a community skill ("uninstall [skill]", "remove this skill") rather than just disable it.

    Also in 6 projects that use this

  • woosikkimSkill

    cancel

    Clear any stale OMC-Slim state

    Also in 3 projects that use this

  • ibahgatSkill

    mode

    Inspect or switch omi operating mode/profile for speed, depth, autonomy, or strict quality gates.

    Also in 3 projects that use this

  • BlockSkill

    open

    Open markdown documents in Penpal for collaborative review. Use when you want a human to review a document, plan, or research artifact.

    Also in 3 projects that use this

  • QinghongLinSkill

    editor

    Read analyst.json and detective.json, make all editorial decisions — what the blog argues, which findings matter, narrative arc and section structure. No visual design. Outputs editor.md (prose) and editor.json (structure with edt_xx IDs).

    Also in 3 projects that use this

  • AnthropicSkill

    status

    Case status summary by audience — client-facing (plain language), internal (for the professor), or court-ready (formal caption format per local rules). Same facts, different framing and depth. Use when a student needs to update the client, brief the professor, or prepare a court status report.

    Also in 15 projects that use this

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 65 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run these and compare with README.md badge/table values:

    line 16
  2. Very specific:

    Check every file referenced in .claude/docs/directory-structure.md:

    line 41
  3. Distinctive:

    After structural changes (adding/removing agents, skills, hooks)

    line 210
  4. Distinctive:

    Before publishing or pushing to GitHub

    line 211
  5. Distinctive:

    After a batch of agent operations (ingest, research, etc.)

    line 212
  6. Distinctive:

    Before starting a new game project from the template

    line 214

“Common” is measured: kept by at least 6 in 10 of the 65 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

/doctor — Repo Health Check

Purpose

One command that catches all drift, broken promises, stale counts, and missing files. The template should never promise what it doesn't ship.

Checks (run all programmatically)

1. Count Accuracy

Run these and compare with README.md badge/table values:

# Actual counts
AGENTS=$(ls .claude/agents/ | wc -l | tr -d ' ')
SKILLS=$(find .claude/skills -name "SKILL.md" | wc -l | tr -d ' ')
HOOKS=$(ls .claude/hooks/ | wc -l | tr -d ' ')
RULES=$(ls .claude/rules/ | wc -l | tr -d ' ')
TEMPLATES=$(ls .claude/docs/templates/ | wc -l | tr -d ' ')
WIKI_PAGES=$(find wiki -name "*.md" -not -path "wiki/raw/*" -type f | wc -l | tr -d ' ')
RAW_SOURCES=$(find wiki/raw -name "*.md" -type f | wc -l | tr -d ' ')

echo "Agents: $AGENTS"
echo "Skills: $SKILLS"
echo "Hooks: $HOOKS"
echo "Rules: $RULES"
echo "Templates: $TEMPLATES"
echo "Wiki pages: $WIKI_PAGES"
echo "Raw sources: $RAW_SOURCES"

Then grep README.md for the claimed numbers and flag any mismatch.

2. Promised Files Exist

Check every file referenced in .claude/docs/directory-structure.md:

# Tool scripts
for f in tools/setup.sh tools/build.sh tools/lint.sh tools/test.sh; do
    [ -f "$f" ] && echo "✅ $f" || echo "❌ $f MISSING"
done

# Toolchain configs
for f in default.project.json selene.toml stylua.toml wally.toml aftman.toml; do
    [ -f "$f" ] && echo "✅ $f" || echo "❌ $f MISSING"
done

# CI
[ -f ".github/workflows/ci.yml" ] && echo "✅ CI workflow" || echo "❌ CI workflow MISSING"

# Production files
for f in production/decision-log.md production/risk-register.md; do
    [ -f "$f" ] && echo "✅ $f" || echo "❌ $f MISSING"
done

# Core docs
for f in CLAUDE.md README.md LICENSE .gitignore UPGRADING.md; do
    [ -f "$f" ] && echo "✅ $f" || echo "❌ $f MISSING"
done

3. Unresolved Placeholders

# Scan for common placeholder patterns
grep -rn "\[CHOOSE:" CLAUDE.md README.md .claude/docs/ 2>/dev/null
grep -rn "\[TODO:" CLAUDE.md README.md 2>/dev/null
grep -rn "\[PLACEHOLDER" CLAUDE.md README.md 2>/dev/null
grep -rn "TBD\|tbd" CLAUDE.md README.md 2>/dev/null | grep -v "^Binary"

Flag any matches as unresolved.

4. Wiki Index Sync

# Compare index page_count with actual disk count
INDEX_COUNT=$(grep "page_count:" wiki/index.md | head -1 | grep -oE '[0-9]+')
DISK_COUNT=$(find wiki -name "*.md" -not -path "wiki/raw/*" -type f | wc -l | tr -d ' ')
echo "Index claims: $INDEX_COUNT"
echo "Disk has: $DISK_COUNT"
[ "$INDEX_COUNT" = "$DISK_COUNT" ] && echo "✅ Synced" || echo "❌ DRIFT: $((DISK_COUNT - INDEX_COUNT)) pages off"

5. Wiki Status Breakdown

for status in complete draft stub; do
    count=$(grep -rl "status: $status" wiki/ --include='*.md' 2>/dev/null | grep -v '/raw/' | wc -l | tr -d ' ')
    echo "$status: $count"
done

6. Toolchain Validity

# JSON validity
python3 -c "import json; json.load(open('default.project.json'))" 2>/dev/null && echo "✅ default.project.json valid JSON" || echo "❌ default.project.json INVALID"

# Key fields
grep -q '"roblox"' selene.toml && echo "✅ selene.toml has std=roblox" || echo "❌ selene.toml missing std"
grep -q 'indent_type' stylua.toml && echo "✅ stylua.toml configured" || echo "❌ stylua.toml empty"
grep -q '\[package\]' wally.toml && echo "✅ wally.toml has package section" || echo "❌ wally.toml missing package"
grep -q '\[tools\]' aftman.toml && echo "✅ aftman.toml has tools" || echo "❌ aftman.toml missing tools"

7. Agent Roster Completeness

# Every .md in agents/ should be mentioned in agent-roster.md
for f in .claude/agents/*.md; do
    name=$(basename "$f" .md)
    if ! grep -q "$name" .claude/docs/agent-roster.md 2>/dev/null; then
        echo "❌ Agent '$name' not in agent-roster.md"
    fi
done
echo "✅ Roster check complete"

8. MCP Integration Check

# Studio MCP agent exists
[ -f ".claude/agents/studio-mcp-operator.md" ] && echo "✅ Studio MCP agent" || echo "❌ Studio MCP agent MISSING"

# Blender MCP agent exists
[ -f ".claude/agents/blender-mcp-operator.md" ] && echo "✅ Blender MCP agent" || echo "❌ Blender MCP agent MISSING"

# CLAUDE.md mentions MCP
grep -q "Studio MCP" CLAUDE.md && echo "✅ CLAUDE.md references Studio MCP" || echo "❌ CLAUDE.md missing Studio MCP ref"
grep -q "Blender MCP" CLAUDE.md && echo "✅ CLAUDE.md references Blender MCP" || echo "❌ CLAUDE.md missing Blender MCP ref"

9. Source File Bootstrap

# Minimum runtime files exist
for f in src/ServerScriptService/Main.server.luau src/StarterPlayer/StarterPlayerScripts/ClientBootstrap.client.luau src/ReplicatedStorage/Shared/Remotes.luau src/ReplicatedStorage/Shared/Types.luau; do
    [ -f "$f" ] && echo "✅ $f" || echo "⚠️ $f not yet created (created by /start)"
done

10. Git State

# Clean working tree
DIRTY=$(git status --porcelain | wc -l | tr -d ' ')
[ "$DIRTY" = "0" ] && echo "✅ Clean working tree" || echo "⚠️ $DIRTY uncommitted changes"

# Up to date with remote
git fetch --dry-run 2>&1 | grep -q "." && echo "⚠️ Remote has new commits" || echo "✅ Up to date"

Output Format

Compile all results into:

# Doctor Report

**Date:** YYYY-MM-DD
**Status:** [HEALTHY / WARNINGS / UNHEALTHY]

## Counts
| Item | README | Actual | Match |
|------|--------|--------|-------|
| Agents | X | X | ✅/❌ |
| Skills | X | X | ✅/❌ |
| Hooks | X | X | ✅/❌ |
| Rules | X | X | ✅/❌ |

## File Integrity
- [✅/❌] All promised files exist
- [List any missing]

## Placeholders
- [✅/❌] No unresolved placeholders

## Wiki
- Index: [synced/drifted]
- Status: X complete, Y draft, Z stub

## Toolchain
- [✅/❌] All configs valid

## Agent Roster
- [✅/❌] All agents in roster

## MCP
- [✅/❌] Studio MCP configured
- [✅/❌] Blender MCP configured

## Git
- [✅/❌] Clean tree
- [✅/❌] Up to date

## Recommendations
1. [Most critical fix]
2. [Next fix]

When to Run

  • After structural changes (adding/removing agents, skills, hooks)
  • Before publishing or pushing to GitHub
  • After a batch of agent operations (ingest, research, etc.)
  • Periodically as a sanity check
  • Before starting a new game project from the template

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.