{"id":50330178,"url":"https://github.com/developerjillur/nexalance-claude-code-kit","last_synced_at":"2026-05-29T09:01:33.162Z","repository":{"id":353982949,"uuid":"1209768549","full_name":"developerjillur/nexalance-claude-code-kit","owner":"developerjillur","description":"AI Development Operating System for Claude Code — v4.4 LITE+ : ~70% fewer tokens vs v4.3, lazy-loaded playbooks, tier-aware Phase 0, risk-tiered review, MemPalace reliability fixes, Graphify integration, hooks-based enforcement","archived":false,"fork":false,"pushed_at":"2026-05-20T18:41:36.000Z","size":1383,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-21T00:16:30.089Z","etag":null,"topics":["ai-coding","ai-development","anthropic","claude","claude-code","claudemd","developer-tools","mempalace","plugins","token-optimization"],"latest_commit_sha":null,"homepage":"","language":"Shell","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/developerjillur.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-04-13T19:03:08.000Z","updated_at":"2026-05-20T18:41:42.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/developerjillur/nexalance-claude-code-kit","commit_stats":null,"previous_names":["developerjillur/nexalance-claude-code-kit"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/developerjillur/nexalance-claude-code-kit","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/developerjillur%2Fnexalance-claude-code-kit","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/developerjillur%2Fnexalance-claude-code-kit/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/developerjillur%2Fnexalance-claude-code-kit/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/developerjillur%2Fnexalance-claude-code-kit/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/developerjillur","download_url":"https://codeload.github.com/developerjillur/nexalance-claude-code-kit/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/developerjillur%2Fnexalance-claude-code-kit/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33644313,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-05-29T02:00:06.066Z","response_time":107,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-coding","ai-development","anthropic","claude","claude-code","claudemd","developer-tools","mempalace","plugins","token-optimization"],"created_at":"2026-05-29T09:01:28.813Z","updated_at":"2026-05-29T09:01:33.156Z","avatar_url":"https://github.com/developerjillur.png","language":"Shell","funding_links":[],"categories":[],"sub_categories":[],"readme":"# NexaLance Claude Code Kit — v4.5 (Auto-Wiki)\n\n\u003e **The complete AI development operating system for Claude Code.**\n\u003e Token-optimized, quality-preserved, production-hardened.\n\n[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)\n[![Claude Code](https://img.shields.io/badge/Claude_Code-Compatible-orange.svg)](https://claude.ai/code)\n[![Version](https://img.shields.io/badge/version-v4.5-success)](https://github.com/developerjillur/nexalance-claude-code-kit/releases)\n[![Token Reduction](https://img.shields.io/badge/tokens-70%25_reduction-brightgreen)](#token-savings)\n[![Plugins](https://img.shields.io/badge/Plugins-8_Integrated-green.svg)](#integrated-plugins)\n\n---\n\n## 🎯 What This Is\n\nA production-ready CLAUDE.md system that solves the biggest pains of running Claude Code on real, large projects:\n\n| Pain | How this kit fixes it |\n|------|------------------------|\n| Claude burns context fast on big projects | **v4.4 LITE: ~70% token reduction**, lazy-loaded playbooks, tiered Phase 0 |\n| Memory lost between sessions | **MemPalace** integration with wing isolation per project |\n| Code review doesn't scale | **Risk-tiered review** — 5-agent for security, inline for routine |\n| Quality drops on multi-task plans | **Deep Focus Mode** + inline self-review checklist |\n| MemPalace \"sometimes\" doesn't work | **`diagnose-mempalace.sh`** + interpreter auto-detection (fixes pyenv/python3 issues) |\n| Mock/stub data sneaks into production | **Zero Mock Data Rule** + connection-chain verification |\n| Browser tests waste tokens | **Playwright CLI** (4× cheaper than MCP) + DOM-targeted snapshots |\n| Multi-project context bleed | **MemPalace Wings** — fully isolated memory per project |\n| Tokens balloon on large codebases | **Optional Graphify** — knowledge-graph queries, ~71× cheaper than re-reading files |\n| No enforcement of token rules | **Runtime hooks** — playbook-tracker, cache-warn, reset-counter |\n| Past prompts/discussions/research scatter and get lost | **Auto-LLM-Wiki** ([Karpathy pattern](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f)) — `docs/wiki/` auto-scaffolds, sessions auto-ingest via hooks, synthesis on demand |\n\n---\n\n## ⭐ What's New in v4.5 (Auto-Wiki)\n\n### NEW: LLM Wiki layer — fully automatic, zero manual operation\n\n[Karpathy's LLM Wiki pattern](https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f) baked into the kit. Every project automatically gets a `docs/wiki/` knowledge base that compounds over time:\n\n- **Auto-ingest hook (zero LLM tokens):** `SessionStart` + `SessionEnd` hooks fire `wiki-ingest.py` which converts Claude Code session JSONL transcripts → markdown into `docs/wiki/raw/discussions/`. Idempotent, ~50ms idle, basic secret redaction (sk-…, ghp_…, AKIA…, Bearer …).\n- **Auto-scaffold:** `setup-project-wing.sh` creates the full `docs/wiki/` structure (CLAUDE.md schema, index.md, log.md, raw/{discussions,prompts,articles,interviews,inbox}, synthesized/{_entities}) at project init.\n- **Auto-backfill:** at end of setup, the hook runs once to convert any pre-existing sessions for that project — wiki starts populated.\n- **On-demand synthesis** (Claude does it, not the user): when `.synthesis-pending` exists OR user asks a domain/recall question, Claude reads `index.md` first, drills 2-4 relevant pages, cites raw sources. See `playbooks/llm-wiki.md`.\n- **Privacy-aware defaults:** `raw/discussions/` and `raw/interviews/` gitignored (may contain secrets). `synthesized/`, `index.md`, `log.md` are committed.\n\n**The user never has to run anything.** Hooks handle plumbing. Claude handles thinking.\n\n### From v4.4 LITE+ (still all in)\n\n#### Token-optimized\n\n- **Slim CLAUDE.md core** (~6K tokens, was ~24K in v4.3) + **12 lazy-loaded playbooks** read only when relevant\n- **Tier-aware Phase 0**: `lite` (3 docs) / `standard` (6, default) / `full` (9), instead of always-9\n- **Risk-tiered review**: inline 5-question check by default, escalates to 5-agent `/code-review` only for security-sensitive / pre-PR\n- **Inline self-review** — saves an LLM round-trip per task; full 7-category fallback still available when escalated\n- **Haiku subagent dispatch** rules for cheap auxiliary tasks (doc updates, status reports, routine commits)\n- **Cache discipline** + Playwright `--interactive-only` snapshot optimization\n\n**Verified token math:** typical session **80–120K → 25–40K** (~65–70% reduction). Quality bar preserved — every v4.3 rule is either in the slim core or in a lazy-loaded playbook.\n\n#### Production-hardened\n\n- **`hooks/` runtime enforcement** (4 scripts auto-installed by setup):\n  - `wiki-ingest.py` — auto-converts session JSONL → wiki markdown (NEW)\n  - `playbook-tracker.sh` — counts playbook reads/turn, warns if \u003e 2\n  - `cache-warn.sh` — warns when CLAUDE.md is edited mid-session (would break prompt cache)\n  - `reset-counter.sh` — resets per-turn counter on each user prompt\n- **Tier-aware Feature Tracker** in SESSION.md for Lite/Standard tiers\n- **MemPalace reliability fixes** — auto-detects working Python 3, uses absolute paths, eliminates silent hook failures\n\n#### Tools\n\n- **`diagnose-mempalace.sh`** — 12-check health diagnostic; pinpoints the exact failing layer with the exact fix command\n- **`migrate-to-v44.sh`** — idempotent v4.3 → v4.x migration; auto-detects existing wing config, full backup, **8-check** verification (NEW: wiki + wiki-ingest checks), safe rollback\n- **`setup-graphify.sh`** — optional Graphify codebase knowledge graph integration (~71× cheaper structural queries on large codebases)\n\n---\n\n## 📦 What's in the Kit\n\n```\nnexalance-claude-code-kit/\n├── NexaLance-CLAUDE-v4.4-LITE.md       Slim core (~6K tokens)\n│\n├── playbooks/                           12 lazy-loaded modules\n│   ├── phase-0.md                       Phase 0 doc generation (tier-aware)\n│   ├── feature-workflow.md              Feature implementation + phase-based dev\n│   ├── testing.md                       Action-Level Testing protocol\n│   ├── browser-automation.md            Playwright CLI + Chrome DevTools\n│   ├── self-review-full.md              7-category fallback (escalation only)\n│   ├── session-management.md            Session start/middle/end + Exit Gate\n│   ├── persistent-memory.md             MemPalace setup + wing protocol\n│   ├── plugin-orchestration.md          Plugin priority + conflict matrix\n│   ├── patterns.md                      Common + anti-patterns\n│   ├── operations.md                    File routing + deployment + rollback\n│   ├── mempalace-troubleshooting.md     Symptom → cause → fix reference\n│   └── llm-wiki.md                      LLM Wiki ingest/query/lint workflow (Karpathy pattern)\n│\n├── hooks/                               4 runtime scripts (auto-installed per project)\n│   ├── wiki-ingest.py                   Auto-converts Claude session JSONL → wiki markdown\n│   ├── playbook-tracker.sh              Counts playbook reads/turn\n│   ├── cache-warn.sh                    Warns on mid-session CLAUDE.md edits\n│   └── reset-counter.sh                 Resets per-turn counter\n│\n├── setup-nexalance.sh                   One-time machine setup (interpreter auto-detection)\n├── setup-project-wing.sh                Per-project setup (tier-aware, wires hooks)\n├── setup-graphify.sh                    Optional Graphify integration\n├── migrate-to-v44.sh                    Safe v4.3 → v4.4 LITE migration\n├── diagnose-mempalace.sh                12-check MemPalace health diagnostic\n├── README.md                            This file\n├── LICENSE                              MIT\n└── .gitignore\n```\n\n---\n\n## 🚀 Quick Start (5 minutes)\n\n### 1. Clone this repo\n\n```bash\ngit clone https://github.com/developerjillur/nexalance-claude-code-kit.git ~/Desktop/nexalance-kit\ncd ~/Desktop/nexalance-kit\nchmod +x *.sh hooks/*.sh\n```\n\n### 2. Run machine setup (once per machine)\n\n```bash\nbash setup-nexalance.sh\n```\n\nInstalls MemPalace + plugins (Superpowers, Code Review, Feature Dev, Frontend Design) + Playwright CLI + Chromium + Chrome DevTools MCP + design tooling. **Auto-detects** your working Python 3 interpreter so MCP and hooks use the right one — no more pyenv/python3 surprises.\n\n### 3. Set up your first project (v4.4 LITE recommended)\n\n```bash\nmkdir -p ~/projects/my-project \u0026\u0026 cd ~/projects/my-project\n\n# Copy the LITE core + playbooks\ncp ~/Desktop/nexalance-kit/NexaLance-CLAUDE-v4.4-LITE.md ./CLAUDE.md\ncp -r ~/Desktop/nexalance-kit/playbooks ./playbooks\n\n# Init project (tier: lite | standard | full — default standard)\nbash ~/Desktop/nexalance-kit/setup-project-wing.sh \\\n    \"my-project\" \"My Awesome Project\" \"client-name\" standard\n\n# Drop your PRD in docs/PRD.md, then open in VS Code\ncode .\n```\n\n### 4. Start building\n\nIn Claude Code panel, type:\n\n```\nStart project\n```\n\nClaude reads `CLAUDE.md` → reads `playbooks/phase-0.md` (lazy-loaded) → generates the docs your tier requires → asks for `Go`. Type **Go**. Building begins.\n\n---\n\n## 📊 Token Savings\n\n| Stage | v4.3 | **v4.4 LITE+** | Saving |\n|-------|------|------------|--------|\n| CLAUDE.md baseline load | ~24K | **~6K** | **−75%** |\n| One playbook on demand | n/a | ~1–3K | (was always-loaded) |\n| Phase 0 (Standard tier, 6 docs) | ~18–30K | **~12–18K** | −33% |\n| Routine feature + inline review | ~25–40K | **~8–15K** | **−65%** |\n| Security feature + 5-agent review | ~30–40K | ~25–35K | preserved (full depth on purpose) |\n| **Typical session total** | **~80–120K** | **~25–40K** | **−65 to −70%** |\n\nQuality bar identical: every rule from v4.3 is either in the slim core (non-negotiables) or a lazy-loaded playbook (deep detail). Nothing removed.\n\n---\n\n## 🪜 Tier System\n\nSet `PROJECT_TIER` once at project init; Phase 0 generates only what's needed.\n\n| Tier | Phase 0 docs | Feature progress lives in | Use for |\n|------|--------------|---------------------------|---------|\n| `lite` | DESIGN, SCHEMA, SESSION (3) | `SESSION.md` → Feature Tracker | MVPs, prototypes, small tools |\n| `standard` *(default)* | + TAD, API, RULES (6) | `SESSION.md` → Feature Tracker | Most production work |\n| `full` | + FEATURES, TASKS, TESTING (9) | Dedicated `FEATURES.md` | Enterprise, multi-team, regulated |\n\nTier flag is just the 4th argument to `setup-project-wing.sh`:\n\n```bash\nbash setup-project-wing.sh \"my-wing\" \"Project\" \"client\" lite     # 3 docs\nbash setup-project-wing.sh \"my-wing\" \"Project\" \"client\" standard # 6 docs (default)\nbash setup-project-wing.sh \"my-wing\" \"Project\" \"client\" full     # 9 docs\n```\n\nNeed a doc your tier didn't generate? Generate it on demand later — no need to re-init.\n\n---\n\n## 🪝 Runtime Hooks (auto-installed)\n\nAuto-installed by `setup-project-wing.sh` into `.claude/hooks/`. Wired into `.claude/settings.json`. They observe, warn, and ingest — they never block.\n\n| Hook | Fires on | What it does |\n|------|----------|--------------|\n| `wiki-ingest.py` ⭐ | `SessionStart`, `SessionEnd` | Converts Claude Code session JSONL → markdown into `docs/wiki/raw/discussions/`. Idempotent, ~50ms idle, secret redaction. |\n| `playbook-tracker.sh` | `Read` of `playbooks/*.md` | Counts reads/turn, warns if \u003e 2 |\n| `cache-warn.sh` | `Edit/Write` of `CLAUDE.md` or playbooks | Warns about prompt-cache invalidation |\n| `reset-counter.sh` | `UserPromptSubmit` | Resets per-turn playbook counter |\n\nDiagnostic files:\n- `.claude/.playbook-counter` — current-turn playbook read count\n- `docs/wiki/.ingest.log` — wiki-ingest history + errors\n- `docs/wiki/.ingest-manifest.json` — processed session IDs (idempotency)\n- `docs/wiki/.synthesis-pending` — flag set when new content awaits synthesis\n\n---\n\n## 📚 Auto-LLM-Wiki (v4.5)\n\nEvery project gets a `docs/wiki/` per-project knowledge base, fully automatic.\n\n### How it runs (zero user action)\n\n1. **`setup-project-wing.sh` scaffolds** the structure at project init:\n   ```\n   docs/wiki/\n   ├── CLAUDE.md           (schema)\n   ├── index.md            (catalog)\n   ├── log.md              (audit)\n   ├── raw/                (immutable: discussions/, prompts/, articles/, interviews/, inbox/)\n   └── synthesized/        (LLM-curated pages + _entities/)\n   ```\n\n2. **Backfill on setup:** `wiki-ingest.py` runs once to convert any pre-existing Claude Code sessions for this project.\n\n3. **Continuous ingest:** `SessionStart` + `SessionEnd` hooks re-run `wiki-ingest.py` every time you open or close Claude Code in this project.\n\n4. **On-demand synthesis (Claude, not the user):** when `.synthesis-pending` exists or you ask a domain/recall question, Claude reads `index.md` first, drills 2–4 relevant pages, cites raw sources back. Workflow: `playbooks/llm-wiki.md`.\n\n### Privacy\n| Path | Committed? | Why |\n|------|-----------|------|\n| `docs/wiki/CLAUDE.md`, `index.md`, `log.md`, `synthesized/**` | ✅ yes | Schema + LLM-cleaned content |\n| `docs/wiki/raw/articles/`, `raw/prompts/` | ✅ yes | User-curated, vetted |\n| `docs/wiki/raw/discussions/`, `raw/interviews/` | ❌ gitignored | May contain secrets / PII |\n| `.ingest-manifest.json`, `.synthesis-pending`, `.ingest.log` | ❌ gitignored | Local state |\n\n### What it complements (not replaces)\n\n| Layer | Stores |\n|-------|--------|\n| MemPalace | Episodic conversation memory (wing-isolated per project) |\n| Graphify *(opt-in)* | Code semantic index (call graph, deps) |\n| **LLM Wiki** | **Compounding domain knowledge for THIS project** |\n| SESSION.md | Current task state |\n\n---\n\n## 🩺 MemPalace Not Working? Run the Diagnostic First\n\nThe single most common issue: `python` (the bare command) doesn't resolve to Python 3 on your machine — common on macOS pyenv users, modern macOS without legacy aliases, and python3-only Linux. Older versions of the kit hardcoded `python -m mempalace.mcp_server`, which silently failed.\n\n**Diagnose any MemPalace problem in 5 seconds:**\n\n```bash\ncd ~/projects/your-project\nbash ~/Desktop/nexalance-kit/diagnose-mempalace.sh\n```\n\nThe script runs **12 checks** across all layers (interpreter, package install, MCP registration, project `.mcp.json`, hooks, `PROJECT_WING` value, wing data, MCP server liveness, and more) and prints **prioritized root causes** with the exact fix command for each.\n\n**Surgical one-liner if you already know it's the python issue:**\n\n```bash\nPY=$(command -v python3)\nclaude mcp remove mempalace --scope user 2\u003e/dev/null\nclaude plugin uninstall mempalace 2\u003e/dev/null\nclaude mcp add mempalace --scope user -- \"$PY\" -m mempalace.mcp_server\nsed -i.bak \"s|python -m mempalace|$PY -m mempalace|g\" ~/.claude/settings.json\n# Restart Claude Code, then verify:\nclaude mcp list | grep mempalace\n```\n\nFor full symptom→cause→fix reference, see [`playbooks/mempalace-troubleshooting.md`](playbooks/mempalace-troubleshooting.md).\n\n**v4.4 LITE+ structural fixes (auto-applied if you re-run setup-nexalance.sh):**\n- Auto-detects working Python 3 interpreter, uses absolute path everywhere\n- Removes duplicate project-level MCP entry\n- Hook errors log to `~/.mempalace-hook.log` instead of `/dev/null`\n- Wing init errors are visible\n- Refuses placeholder wing names (`\"project-wing-name\"`)\n- Wing seeded with 4 triples on init so `mempalace_search` returns data immediately\n\n---\n\n## 🔄 Migrating an Existing Project from v4.3 → v4.4 LITE\n\n**One command (recommended):**\n\n```bash\ncd ~/projects/your-existing-project\nbash ~/Desktop/nexalance-kit/migrate-to-v44.sh standard   # or: lite | full\n```\n\nAuto-detects existing `PROJECT_WING`, `PROJECT_NAME`, `CLIENT`. Backs up to `.nexalance-backup-\u003ctimestamp\u003e/`. Installs LITE core + 11 playbooks + 3 hooks. Upgrades `SESSION.md` to add Feature Tracker for Lite/Standard tiers (preserves history). Runs **6 verification checks** before declaring success. **Idempotent** — re-running on an already-migrated project aborts safely.\n\n**Rollback (paths in script output):**\n\n```bash\ncp .nexalance-backup-\u003ctimestamp\u003e/CLAUDE.md.v43 CLAUDE.md\ncp .nexalance-backup-\u003ctimestamp\u003e/settings.json.v43 .claude/settings.json\nrm -rf playbooks/ .claude/hooks/\n```\n\n---\n\n## 🕸️ Optional Add-On: Graphify (Codebase Knowledge Graph)\n\nFor **large or multi-language codebases**, add Graphify — a knowledge-graph index over your code, docs, papers, and images. Claude queries the graph instead of re-reading files (~71× fewer tokens per structural query).\n\n**Layer model:**\n\n| Layer | Tool | What it stores |\n|-------|------|----------------|\n| Episodic (\"what we did\") | **MemPalace** | conversation memory, decisions, per-project wings |\n| Semantic (\"what the code IS\") | **Graphify** | structure, deps, call graph, doc concepts |\n| Session state | **SESSION.md / HANDOFF.md** | current task, next step |\n\n### Install (once per machine)\n\n```bash\nbash ~/Desktop/nexalance-kit/setup-graphify.sh\n```\n\n### Use it (per project)\n\n```bash\n# In Claude Code, after Phase 0:\n/graphify .\n\n# Common queries:\n/graphify query \"how does the auth flow work?\"\n/graphify path \"LoginForm\" \"sessionStore\"\n/graphify .  --update                       # incremental refresh\n/graphify clone https://github.com/foo/bar  # index an external repo\n/graphify merge-graphs g1.json g2.json      # cross-repo graph\n```\n\n**When to add it:** ✅ Codebase \u003e30k LOC, multi-language, lots of `/docs` PDFs/specs. ❌ Skip for tiny projects, greenfield/empty repos, or pure PRD work.\n\nCLAUDE.md core has a rule: query `graph.json` BEFORE Glob/Grep on \u003e20 files when `graphify-out/` exists. Auto-applied.\n\n---\n\n## 🔄 Daily Workflow\n\n### Start working\n\n```\n1. Open VS Code with your project folder\n2. Open Claude Code panel (Spark icon)\n3. Type: \"Continue\" or \"Resume\"\n4. Claude loads MemPalace memory → reads SESSION.md → resumes the exact task\n```\n\n### Switch projects\n\n```\n1. Open a different project folder\n2. Open Claude Code panel\n3. Type: \"Continue\"\n4. MemPalace switches to that project's wing automatically — zero context bleed\n```\n\n### End your day\n\n```\n1. Type: \"I'm done for today\"\n2. Claude automatically:\n   → Runs Session Exit Gate (codebase review + tests)\n   → Runs risk-tiered review (5-agent if pre-PR; otherwise inline)\n   → Saves to MemPalace\n   → Updates SESSION.md + HANDOFF.md\n   → Git commits all progress\n```\n\n### Natural commands (no plugin syntax to remember)\n\n```\n\"Build the user auth system\"   → /feature-dev auto-runs\n\"Test the login form\"           → Playwright CLI opens browser\n\"Why is this page broken?\"      → Chrome DevTools checks console\n\"Review the code\"               → tier-appropriate review (inline or 5-agent)\n\"Plan the payment integration\"  → Superpowers brainstorm\n\"Deploy to hosting\"             → Playwright CLI headed browser\n```\n\n---\n\n## 🔌 Integrated Plugins\n\n| Plugin | Purpose | Auto-trigger |\n|--------|---------|--------------|\n| **Superpowers** | Brainstorm → Plan → TDD → Execute | New ideas / planning |\n| **MemPalace** | Persistent memory across sessions | Every session start/end |\n| **Code Review** | 5-agent parallel review | Security-sensitive or pre-PR |\n| **Feature Dev** | 7-phase feature implementation | Complex features |\n| **Frontend Design** | Anti-AI-slop design quality | Any frontend work |\n| **Playwright CLI** | Browser testing (4× cheaper than MCP) | Testing, deploying |\n| **Chrome DevTools** | Console, network, performance debugging | UI bugs |\n| **Vercel Guidelines** | 100+ UX/accessibility rules | UI quality checks |\n| **Graphify** *(opt-in)* | Codebase knowledge graph | Big-codebase exploration |\n\n---\n\n## 🏗️ Multi-Project Setup\n\nEach project gets its own MemPalace \"wing\" — fully isolated memory:\n\n```bash\n# Project 1 (Standard tier)\ncd ~/projects/project-alpha\ncp ~/Desktop/nexalance-kit/NexaLance-CLAUDE-v4.4-LITE.md ./CLAUDE.md\ncp -r ~/Desktop/nexalance-kit/playbooks ./playbooks\nbash ~/Desktop/nexalance-kit/setup-project-wing.sh \"alpha\" \"Project Alpha\" \"client-a\" standard\n\n# Project 2 (Full tier — enterprise)\ncd ~/projects/project-beta\ncp ~/Desktop/nexalance-kit/NexaLance-CLAUDE-v4.4-LITE.md ./CLAUDE.md\ncp -r ~/Desktop/nexalance-kit/playbooks ./playbooks\nbash ~/Desktop/nexalance-kit/setup-project-wing.sh \"beta\" \"Project Beta\" \"client-b\" full\n\n# Switching: just open the folder → Claude loads correct wing\n```\n\n---\n\n## ❓ Troubleshooting\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eMemPalace shows \"Failed to connect\"\u003c/strong\u003e\u003c/summary\u003e\n\n```bash\nbash ~/Desktop/nexalance-kit/diagnose-mempalace.sh\n```\n\nThe script identifies the exact root cause and prints the fix command. Most common: the kit's older configs hardcoded `python` which fails on pyenv / python3-only systems. Re-running `setup-nexalance.sh` from the v4.4 LITE+ release auto-fixes this.\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003ePlaywright CLI \"not found\"\u003c/strong\u003e\u003c/summary\u003e\n\n```bash\nnpm install -g @playwright/cli\nnpx playwright install chromium\n```\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003ePlugin not installed\u003c/strong\u003e\u003c/summary\u003e\n\nIn Claude Code:\n```\n/plugin marketplace add obra/superpowers-marketplace\n/plugin install superpowers@superpowers-marketplace\n/plugin install code-review@claude-plugins-official\n/plugin install feature-dev@claude-plugins-official\n/plugin install frontend-design@claude-plugins-official\n```\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eContext getting heavy / slow responses\u003c/strong\u003e\u003c/summary\u003e\n\nIn Claude Code: `/compact`. Or start a new session — MemPalace preserves all context. Also check `.claude/.playbook-counter` — if \u003e 2, you've been reading too many playbooks per turn (review the routing table in CLAUDE.md core).\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eBrowser test needs login\u003c/strong\u003e\u003c/summary\u003e\n\nTell Claude: \"Open browser and navigate to [URL]\". Browser opens (headed mode); you log in manually; tell Claude: \"Continue testing\".\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eNew session lost context\u003c/strong\u003e\u003c/summary\u003e\n\nType: \"Read docs/HANDOFF.md and docs/SESSION.md, then resume.\" With MemPalace properly working, this happens automatically. If it doesn't → run the diagnostic.\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eWing memory bleeding between projects\u003c/strong\u003e\u003c/summary\u003e\n\nEvery MemPalace operation MUST include `--wing PROJECT_WING`. Check that `PROJECT_WING` in each project's CLAUDE.md is unique. Audit recent calls: search for `mempalace_search` in your transcript and verify each has `--wing`.\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eHook errors flooding `~/.mempalace-hook.log`\u003c/strong\u003e\u003c/summary\u003e\n\nThis file is good news — it means hooks are firing and you can see what's failing. Common entries:\n\n- `ModuleNotFoundError: No module named 'mempalace'` → re-run `setup-nexalance.sh` (interpreter mismatch)\n- `pyenv: python: command not found` → same fix\n- `Permission denied` → `chmod -R u+w ~/.mempalace`\n\n\u003c/details\u003e\n\n---\n\n## 📈 Version History\n\n| Version | Lines | Score | Key change |\n|---------|-------|-------|------------|\n| v2.0 | 415 | 7.2 | Basic orchestrator |\n| v3.0 | 913 | 9.1 | Security, deployment, rollback |\n| v4.0 | 1,409 | 9.6 | Deep focus, action testing, self-review, memory |\n| v4.1 | 1,641 | 9.6 | Superpowers + MemPalace + Official plugins |\n| v4.2 | 1,793 | 8.4 | Playwright CLI + Chrome DevTools |\n| v4.3 | 2,350+ | 9.2 | Design system, conflict resolution, file organization |\n| v4.4 LITE+ | core ~530 + playbooks ~2,000 | 9.0 | Token-optimized: slim core + lazy-loaded playbooks, tiered Phase 0, risk-tiered review, MemPalace reliability fixes |\n| **v4.5** | **+ docs/wiki/ auto-scaffold + wiki-ingest hook + llm-wiki playbook** | **9.2** | **Auto-LLM-Wiki: Karpathy pattern fully automated. Every project auto-scaffolds `docs/wiki/`, hooks auto-convert session JSONL → markdown, on-demand synthesis. Zero manual user action after setup.** |\n\n---\n\n## 🏢 Credits\n\nBuilt by **NexaLance Agency** — a full-service digital agency delivering 500+ projects across 30+ countries since 2016.\n\n- Website: [nexalance.com](https://nexalance.com)\n- GitHub: [@developerjillur](https://github.com/developerjillur)\n- Email: info@sorobindu.com\n\nEngineering co-pilot: Claude Opus 4.7 (1M context).\n\n---\n\n## 📄 License\n\nMIT License — free to use, modify, and distribute.\n\n---\n\n*NexaLance Agency — Your Success, Our Mission.*\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeveloperjillur%2Fnexalance-claude-code-kit","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdeveloperjillur%2Fnexalance-claude-code-kit","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeveloperjillur%2Fnexalance-claude-code-kit/lists"}