Merge pull request #986 from Infiniteyieldai/claude/evaluate-repo-comparison-ASZ9Y

docs: Add repo assessment, commands reference, and evaluation docs
2026-03-30 13:43:26 +08:00 · 2026-03-28 20:04:53 -04:00
parent a3fc90f7ac c865d4c676
commit 64847d0a21
4 changed files with 524 additions and 0 deletions
--- a/.claude/rules/node.md
+++ b/.claude/rules/node.md
@@ -0,0 +1,47 @@
+# Node.js Rules for everything-claude-code
+
+> Project-specific rules for the ECC codebase. Extends common rules.
+
+## Stack
+
+- **Runtime**: Node.js >=18 (no transpilation, plain CommonJS)
+- **Test runner**: `node tests/run-all.js` — individual files via `node tests/**/*.test.js`
+- **Linter**: ESLint (`@eslint/js`, flat config)
+- **Coverage**: c8
+- **Lint**: markdownlint-cli for `.md` files
+
+## File Conventions
+
+- `scripts/` — Node.js utilities, hooks. CommonJS (`require`/`module.exports`)
+- `agents/`, `commands/`, `skills/`, `rules/` — Markdown with YAML frontmatter
+- `tests/` — Mirror the `scripts/` structure. Test files named `*.test.js`
+- File naming: **lowercase with hyphens** (e.g. `session-start.js`, `post-edit-format.js`)
+
+## Code Style
+
+- CommonJS only — no ESM (`import`/`export`) unless file ends in `.mjs`
+- No TypeScript — plain `.js` throughout
+- Prefer `const` over `let`; never `var`
+- Keep hook scripts under 200 lines — extract helpers to `scripts/lib/`
+- All hooks must `exit 0` on non-critical errors (never block tool execution unexpectedly)
+
+## Hook Development
+
+- Hook scripts normally receive JSON on stdin, but hooks routed through `scripts/hooks/run-with-flags.js` can export `run(rawInput)` and let the wrapper handle parsing/gating
+- Async hooks: mark `"async": true` in `settings.json` with a timeout ≤30s
+- Blocking hooks (PreToolUse, stop): keep fast (<200ms) — no network calls
+- Use `run-with-flags.js` wrapper for all hooks so `ECC_HOOK_PROFILE` and `ECC_DISABLED_HOOKS` runtime gating works
+- Always exit 0 on parse errors; log to stderr with `[HookName]` prefix
+
+## Testing Requirements
+
+- Run `node tests/run-all.js` before committing
+- New scripts in `scripts/lib/` require a matching test in `tests/lib/`
+- New hooks require at least one integration test in `tests/hooks/`
+
+## Markdown / Agent Files
+
+- Agents: YAML frontmatter with `name`, `description`, `tools`, `model`
+- Skills: sections — When to Use, How It Works, Examples
+- Commands: `description:` frontmatter line required
+- Run `npx markdownlint-cli '**/*.md' --ignore node_modules` before committing
--- a/COMMANDS-QUICK-REF.md
+++ b/COMMANDS-QUICK-REF.md
@@ -0,0 +1,159 @@
+# Commands Quick Reference
+
+> 59 slash commands installed globally. Type `/` in any Claude Code session to invoke.
+
+---
+
+## Core Workflow
+
+| Command | What it does |
+|---------|-------------|
+| `/plan` | Restate requirements, assess risks, write step-by-step implementation plan — **waits for your confirm before touching code** |
+| `/tdd` | Enforce test-driven development: scaffold interface → write failing test → implement → verify 80%+ coverage |
+| `/code-review` | Full code quality, security, and maintainability review of changed files |
+| `/build-fix` | Detect and fix build errors — delegates to the right build-resolver agent automatically |
+| `/verify` | Run the full verification loop: build → lint → test → type-check |
+| `/quality-gate` | Quality gate check against project standards |
+
+---
+
+## Testing
+
+| Command | What it does |
+|---------|-------------|
+| `/tdd` | Universal TDD workflow (any language) |
+| `/e2e` | Generate + run Playwright end-to-end tests, capture screenshots/videos/traces |
+| `/test-coverage` | Report test coverage, identify gaps |
+| `/go-test` | TDD workflow for Go (table-driven, 80%+ coverage with `go test -cover`) |
+| `/kotlin-test` | TDD for Kotlin (Kotest + Kover) |
+| `/rust-test` | TDD for Rust (cargo test, integration tests) |
+| `/cpp-test` | TDD for C++ (GoogleTest + gcov/lcov) |
+
+---
+
+## Code Review
+
+| Command | What it does |
+|---------|-------------|
+| `/code-review` | Universal code review |
+| `/python-review` | Python — PEP 8, type hints, security, idiomatic patterns |
+| `/go-review` | Go — idiomatic patterns, concurrency safety, error handling |
+| `/kotlin-review` | Kotlin — null safety, coroutine safety, clean architecture |
+| `/rust-review` | Rust — ownership, lifetimes, unsafe usage |
+| `/cpp-review` | C++ — memory safety, modern idioms, concurrency |
+
+---
+
+## Build Fixers
+
+| Command | What it does |
+|---------|-------------|
+| `/build-fix` | Auto-detect language and fix build errors |
+| `/go-build` | Fix Go build errors and `go vet` warnings |
+| `/kotlin-build` | Fix Kotlin/Gradle compiler errors |
+| `/rust-build` | Fix Rust build + borrow checker issues |
+| `/cpp-build` | Fix C++ CMake and linker problems |
+| `/gradle-build` | Fix Gradle errors for Android / KMP |
+
+---
+
+## Planning & Architecture
+
+| Command | What it does |
+|---------|-------------|
+| `/plan` | Implementation plan with risk assessment |
+| `/multi-plan` | Multi-model collaborative planning |
+| `/multi-workflow` | Multi-model collaborative development |
+| `/multi-backend` | Backend-focused multi-model development |
+| `/multi-frontend` | Frontend-focused multi-model development |
+| `/multi-execute` | Multi-model collaborative execution |
+| `/orchestrate` | Guide for tmux/worktree multi-agent orchestration |
+| `/devfleet` | Orchestrate parallel Claude Code agents via DevFleet |
+
+---
+
+## Session Management
+
+| Command | What it does |
+|---------|-------------|
+| `/save-session` | Save current session state to `~/.claude/session-data/` |
+| `/resume-session` | Load the most recent saved session from the canonical session store and resume from where you left off |
+| `/sessions` | Browse, search, and manage session history with aliases from `~/.claude/session-data/` (with legacy reads from `~/.claude/sessions/`) |
+| `/checkpoint` | Mark a checkpoint in the current session |
+| `/aside` | Answer a quick side question without losing current task context |
+| `/context-budget` | Analyse context window usage — find token overhead, optimise |
+
+---
+
+## Learning & Improvement
+
+| Command | What it does |
+|---------|-------------|
+| `/learn` | Extract reusable patterns from the current session |
+| `/learn-eval` | Extract patterns + self-evaluate quality before saving |
+| `/evolve` | Analyse learned instincts, suggest evolved skill structures |
+| `/promote` | Promote project-scoped instincts to global scope |
+| `/instinct-status` | Show all learned instincts (project + global) with confidence scores |
+| `/instinct-export` | Export instincts to a file |
+| `/instinct-import` | Import instincts from a file or URL |
+| `/skill-create` | Analyse local git history → generate a reusable skill |
+| `/skill-health` | Skill portfolio health dashboard with analytics |
+| `/rules-distill` | Scan skills, extract cross-cutting principles, distill into rules |
+
+---
+
+## Refactoring & Cleanup
+
+| Command | What it does |
+|---------|-------------|
+| `/refactor-clean` | Remove dead code, consolidate duplicates, clean up structure |
+| `/prompt-optimize` | Analyse a draft prompt and output an optimised ECC-enriched version |
+
+---
+
+## Docs & Research
+
+| Command | What it does |
+|---------|-------------|
+| `/docs` | Look up current library/API documentation via Context7 |
+| `/update-docs` | Update project documentation |
+| `/update-codemaps` | Regenerate codemaps for the codebase |
+
+---
+
+## Loops & Automation
+
+| Command | What it does |
+|---------|-------------|
+| `/loop-start` | Start a recurring agent loop on an interval |
+| `/loop-status` | Check status of running loops |
+| `/claw` | Start NanoClaw v2 — persistent REPL with model routing, skill hot-load, branching, and metrics |
+
+---
+
+## Project & Infrastructure
+
+| Command | What it does |
+|---------|-------------|
+| `/projects` | List known projects and their instinct statistics |
+| `/harness-audit` | Audit the agent harness configuration for reliability and cost |
+| `/eval` | Run the evaluation harness |
+| `/model-route` | Route a task to the right model (Haiku / Sonnet / Opus) |
+| `/pm2` | PM2 process manager initialisation |
+| `/setup-pm` | Configure package manager (npm / pnpm / yarn / bun) |
+
+---
+
+## Quick Decision Guide
+
+```
+Starting a new feature?         → /plan first, then /tdd
+Code just written?              → /code-review
+Build broken?                   → /build-fix
+Need live docs?                 → /docs <library>
+Session about to end?           → /save-session or /learn-eval
+Resuming next day?              → /resume-session
+Context getting heavy?          → /context-budget then /checkpoint
+Want to extract what you learned? → /learn-eval then /evolve
+Running repeated tasks?         → /loop-start
+```
--- a/EVALUATION.md
+++ b/EVALUATION.md
@@ -0,0 +1,122 @@
+# Repo Evaluation vs Current Setup
+
+**Date:** 2026-03-21
+**Branch:** `claude/evaluate-repo-comparison-ASZ9Y`
+
+---
+
+## Current Setup (`~/.claude/`)
+
+The active Claude Code installation is near-minimal:
+
+| Component | Current |
+|-----------|---------|
+| Agents | 0 |
+| Skills | 0 installed |
+| Commands | 0 |
+| Hooks | 1 (Stop: git check) |
+| Rules | 0 |
+| MCP configs | 0 |
+
+**Installed hooks:**
+- `Stop` → `stop-hook-git-check.sh` — blocks session end if there are uncommitted changes or unpushed commits
+
+**Installed permissions:**
+- `Skill` — allows skill invocations
+
+**Plugins:** Only `blocklist.json` (no active plugins installed)
+
+---
+
+## This Repo (`everything-claude-code` v1.9.0)
+
+| Component | Repo |
+|-----------|------|
+| Agents | 28 |
+| Skills | 116 |
+| Commands | 59 |
+| Rules sets | 12 languages + common (60+ rule files) |
+| Hooks | Comprehensive system (PreToolUse, PostToolUse, SessionStart, Stop) |
+| MCP configs | 1 (Context7 + others) |
+| Schemas | 9 JSON validators |
+| Scripts/CLI | 46+ Node.js modules + multiple CLIs |
+| Tests | 58 test files |
+| Install profiles | core, developer, security, research, full |
+| Supported harnesses | Claude Code, Codex, Cursor, OpenCode |
+
+---
+
+## Gap Analysis
+
+### Hooks
+- **Current:** 1 Stop hook (git hygiene check)
+- **Repo:** Full hook matrix covering:
+  - Dangerous command blocking (`rm -rf`, force pushes)
+  - Auto-formatting on file edits
+  - Dev server tmux enforcement
+  - Cost tracking
+  - Session evaluation and governance capture
+  - MCP health monitoring
+
+### Agents (28 missing)
+The repo provides specialized agents for every major workflow:
+- Language reviewers: TypeScript, Python, Go, Java, Kotlin, Rust, C++, Flutter
+- Build resolvers: Go, Java, Kotlin, Rust, C++, PyTorch
+- Workflow agents: planner, tdd-guide, code-reviewer, security-reviewer, architect
+- Automation: loop-operator, doc-updater, refactor-cleaner, harness-optimizer
+
+### Skills (116 missing)
+Domain knowledge modules covering:
+- Language patterns (Python, Go, Kotlin, Rust, C++, Java, Swift, Perl, Laravel, Django)
+- Testing strategies (TDD, E2E, coverage)
+- Architecture patterns (backend, frontend, API design, database migrations)
+- AI/ML workflows (Claude API, eval harness, agent loops, cost-aware pipelines)
+- Business workflows (investor materials, market research, content engine)
+
+### Commands (59 missing)
+- `/tdd`, `/plan`, `/e2e`, `/code-review` — core dev workflows
+- `/sessions`, `/save-session`, `/resume-session` — session persistence
+- `/orchestrate`, `/multi-plan`, `/multi-execute` — multi-agent coordination
+- `/learn`, `/skill-create`, `/evolve` — continuous improvement
+- `/build-fix`, `/verify`, `/quality-gate` — build/quality automation
+
+### Rules (60+ files missing)
+Language-specific coding style, patterns, testing, and security guidelines for:
+TypeScript, Python, Go, Java, Kotlin, Rust, C++, C#, Swift, Perl, PHP, and common/cross-language rules.
+
+---
+
+## Recommendations
+
+### Immediate value (core install)
+Run `ecc install --profile core` to get:
+- Core agents (code-reviewer, planner, tdd-guide, security-reviewer)
+- Essential skills (tdd-workflow, coding-standards, security-review)
+- Key commands (/tdd, /plan, /code-review, /build-fix)
+
+### Full install
+Run `ecc install --profile full` to get all 28 agents, 116 skills, and 59 commands.
+
+### Hooks upgrade
+The current Stop hook is solid. The repo's `hooks.json` adds:
+- Dangerous command blocking (safety)
+- Auto-formatting (quality)
+- Cost tracking (observability)
+- Session evaluation (learning)
+
+### Rules
+Adding language rules (e.g., TypeScript, Python) provides always-on coding guidelines without relying on per-session prompts.
+
+---
+
+## What the Current Setup Does Well
+
+- The `stop-hook-git-check.sh` Stop hook is production-quality and already enforces good git hygiene
+- The `Skill` permission is correctly configured
+- The setup is clean with no conflicts or cruft
+
+---
+
+## Summary
+
+The current setup is essentially a blank slate with one well-implemented git hygiene hook. This repo provides a complete, production-tested enhancement layer covering agents, skills, commands, hooks, and rules — with a selective install system so you can add exactly what you need without bloating the configuration.
--- a/REPO-ASSESSMENT.md
+++ b/REPO-ASSESSMENT.md
@@ -0,0 +1,196 @@
+# Repo & Fork Assessment + Setup Recommendations
+
+**Date:** 2026-03-21
+
+---
+
+## What's Available
+
+### Repo: `Infiniteyieldai/everything-claude-code`
+
+This is a **fork of `affaan-m/everything-claude-code`** (the upstream project with 50K+ stars, 6K+ forks).
+
+| Attribute | Value |
+|-----------|-------|
+| Version | 1.9.0 (current) |
+| Status | Clean fork — 1 commit ahead of upstream `main` (the EVALUATION.md doc added in this session) |
+| Remote branches | `main`, `claude/evaluate-repo-comparison-ASZ9Y` |
+| Upstream sync | Fully synced — last upstream commit merged was the zh-CN docs PR (#728) |
+| License | MIT |
+
+**This is the right repo to work from.** It's the latest upstream version with no divergence or merge conflicts.
+
+---
+
+### Current `~/.claude/` Installation
+
+| Component | Installed | Available in Repo |
+|-----------|-----------|-------------------|
+| Agents | 0 | 28 |
+| Skills | 0 | 116 |
+| Commands | 0 | 59 |
+| Rules | 0 | 60+ files (12 languages) |
+| Hooks | 1 (git Stop check) | Full PreToolUse/PostToolUse matrix |
+| MCP configs | 0 | 1 (Context7) |
+
+The existing Stop hook (`stop-hook-git-check.sh`) is solid — blocks session end on uncommitted/unpushed work. Keep it.
+
+---
+
+## Install Profile Recommendations
+
+The repo ships 5 install profiles. Choose based on your primary use case:
+
+### Profile: `core` (Minimum viable setup)
+> Fastest to install. Gets you commands, core agents, hooks runtime, and quality workflow.
+
+**Best for:** Trying ECC out, minimal footprint, or a constrained environment.
+
+```bash
+node scripts/install-plan.js --profile core
+node scripts/install-apply.js
+```
+
+**Installs:** rules-core, agents-core, commands-core, hooks-runtime, platform-configs, workflow-quality
+
+---
+
+### Profile: `developer` (Recommended for daily dev work)
+> The default engineering profile for most ECC users.
+
+**Best for:** General software development across app codebases.
+
+```bash
+node scripts/install-plan.js --profile developer
+node scripts/install-apply.js
+```
+
+**Adds over core:** framework-language skills, database patterns, orchestration commands
+
+---
+
+### Profile: `security`
+> Baseline runtime + security-specific agents and rules.
+
+**Best for:** Security-focused workflows, code audits, vulnerability reviews.
+
+---
+
+### Profile: `research`
+> Investigation, synthesis, and publishing workflows.
+
+**Best for:** Content creation, investor materials, market research, cross-posting.
+
+---
+
+### Profile: `full`
+> Everything — all 18 modules.
+
+**Best for:** Power users who want the complete toolkit.
+
+```bash
+node scripts/install-plan.js --profile full
+node scripts/install-apply.js
+```
+
+---
+
+## Priority Additions (High Value, Low Risk)
+
+Regardless of profile, these components add immediate value:
+
+### 1. Core Agents (highest ROI)
+
+| Agent | Why it matters |
+|-------|----------------|
+| `planner.md` | Breaks complex tasks into implementation plans |
+| `code-reviewer.md` | Quality and maintainability review |
+| `tdd-guide.md` | TDD workflow (RED→GREEN→IMPROVE) |
+| `security-reviewer.md` | Vulnerability detection |
+| `architect.md` | System design & scalability decisions |
+
+### 2. Key Commands
+
+| Command | Why it matters |
+|---------|----------------|
+| `/plan` | Implementation planning before coding |
+| `/tdd` | Test-driven workflow |
+| `/code-review` | On-demand review |
+| `/build-fix` | Automated build error resolution |
+| `/learn` | Extract patterns from current session |
+
+### 3. Hook Upgrades (from `hooks/hooks.json`)
+The repo's hook system adds these over the current single Stop hook:
+
+| Hook | Trigger | Value |
+|------|---------|-------|
+| `block-no-verify` | PreToolUse: Bash | Blocks `--no-verify` git flag abuse |
+| `pre-bash-git-push-reminder` | PreToolUse: Bash | Pre-push review reminder |
+| `doc-file-warning` | PreToolUse: Write | Warns on non-standard doc files |
+| `suggest-compact` | PreToolUse: Edit/Write | Suggests compaction at logical intervals |
+| Continuous learning observer | PreToolUse: * | Captures tool use patterns for skill improvement |
+
+### 4. Rules (Always-on guidelines)
+The `rules/common/` directory provides baseline guidelines that fire on every session:
+- `security.md` — Security guardrails
+- `testing.md` — 80%+ coverage requirement
+- `git-workflow.md` — Conventional commits, branch strategy
+- `coding-style.md` — Cross-language style standards
+
+---
+
+## What to Do With the Fork
+
+### Option A: Use as upstream tracker (current state)
+Keep the fork synced with `affaan-m/everything-claude-code` upstream. Periodically merge upstream changes:
+```bash
+git fetch upstream
+git merge upstream/main
+```
+Install from the local clone. This is clean and maintainable.
+
+### Option B: Customize the fork
+Add personal skills, agents, or commands to the fork. Good for:
+- Business-specific domain skills (your vertical)
+- Team-specific coding conventions
+- Custom hooks for your stack
+
+The fork already has the EVALUATION.md and REPO-ASSESSMENT.md docs — that's fine for a working fork.
+
+### Option C: Install from npm (simplest for fresh machines)
+```bash
+npx ecc-universal install --profile developer
+```
+No need to clone the repo. This is the recommended install method for most users.
+
+---
+
+## Recommended Setup Steps
+
+1. **Keep the existing Stop hook** — it's doing its job
+2. **Run the developer profile install** from the local fork:
+   ```bash
+   cd /path/to/everything-claude-code
+   node scripts/install-plan.js --profile developer
+   node scripts/install-apply.js
+   ```
+3. **Add language rules** for your primary stack (TypeScript, Python, Go, etc.):
+   ```bash
+   node scripts/install-plan.js --add rules/typescript
+   node scripts/install-apply.js
+   ```
+4. **Enable MCP Context7** for live documentation lookup:
+   - Copy `mcp-configs/mcp-servers.json` into your project's `.claude/` dir
+5. **Review hooks** — enable the `hooks/hooks.json` additions selectively, starting with `block-no-verify` and `pre-bash-git-push-reminder`
+
+---
+
+## Summary
+
+| Question | Answer |
+|----------|--------|
+| Is the fork healthy? | Yes — fully synced with upstream v1.9.0 |
+| Other forks to consider? | None visible in this environment; upstream `affaan-m/everything-claude-code` is the source of truth |
+| Best install profile? | `developer` for day-to-day dev work |
+| Biggest gap in current setup? | 0 agents installed — add at minimum: planner, code-reviewer, tdd-guide, security-reviewer |
+| Quickest win? | Run `node scripts/install-plan.js --profile core && node scripts/install-apply.js` |