continuous-improvement 3.0.0 → 3.1.0

package/CHANGELOG.md

@@ -4,6 +4,25 @@ All notable changes to this skill are documented here.
 
 ---
 
+## [3.0.0] — 2026-04-06
+
+### Added
+- **MCP server** (`bin/mcp-server.mjs`) — zero-dependency JSON-RPC stdio server exposing instincts as MCP tools and resources. Works with Claude Code, Claude Desktop, Cursor, Zed, Windsurf, VS Code.
+- **Beginner / Expert modes** — `--mode beginner` (default, 3 tools) vs `--mode expert` (8 tools with import/export, observation viewer, manual instinct creation, confidence tuning)
+- **MCP-only mode** — `--mode mcp` for editors that support MCP but not Claude Code hooks
+- **Plugin manifests** — `plugins/beginner.json` and `plugins/expert.json` describe available tools per mode
+- **Session hooks** (`hooks/session.sh`) — SessionStart loads instincts and shows status; SessionEnd reminds to reflect
+- **Import/export tools** — `ci_export` and `ci_import` for sharing instincts as JSON between team members
+- **34-test suite** — added MCP server tests (beginner + expert mode) and plugin config validation. Up from 20 tests.
+- **Multi-editor MCP support** — installer patches both `settings.json` and `claude_desktop_config.json`
+
+### Changed
+- **Installer upgraded** — `--mode` flag replaces single-mode install. Supports `beginner`, `expert`, `mcp`
+- **Uninstaller upgraded** — cleans up MCP server config, session hooks, and desktop config
+- **Version bump** to 3.0.0 — breaking change: new install modes and MCP server architecture
+
+---
+
 ## [2.3.0] — 2026-04-06
 
 ### Changed

package/README.md

@@ -11,8 +11,11 @@
 <p align="center">
   <a href="https://www.npmjs.com/package/continuous-improvement"><img src="https://img.shields.io/npm/v/continuous-improvement" alt="npm"></a>
   <a href="https://www.npmjs.com/package/continuous-improvement"><img src="https://img.shields.io/npm/dm/continuous-improvement" alt="downloads"></a>
+  <a href="https://github.com/naimkatiman/continuous-improvement/stargazers"><img src="https://img.shields.io/github/stars/naimkatiman/continuous-improvement?style=social" alt="stars"></a>
+  <a href="https://github.com/naimkatiman/continuous-improvement/network/members"><img src="https://img.shields.io/github/forks/naimkatiman/continuous-improvement?style=social" alt="forks"></a>
+  <a href="https://github.com/naimkatiman/continuous-improvement/graphs/contributors"><img src="https://img.shields.io/github/contributors/naimkatiman/continuous-improvement" alt="contributors"></a>
   <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-green" alt="license"></a>
-  <a href="test/"><img src="https://img.shields.io/badge/tests-20%20passing-brightgreen" alt="tests"></a>
+  <a href="test/"><img src="https://img.shields.io/badge/tests-104%20passing-brightgreen" alt="tests"></a>
 </p>
 
 <p align="center">
@@ -20,6 +23,12 @@
   <a href="https://cursor.sh"><img src="https://img.shields.io/badge/Cursor-compatible-blue" alt="Cursor"></a>
   <a href="https://openai.com/codex"><img src="https://img.shields.io/badge/Codex-compatible-blue" alt="Codex"></a>
   <a href="https://ai.google.dev/gemini-api/docs/gemini-cli"><img src="https://img.shields.io/badge/Gemini%20CLI-compatible-blue" alt="Gemini CLI"></a>
+  <a href="https://github.com/naimkatiman/continuous-improvement/actions"><img src="https://img.shields.io/github/actions/workflow/status/naimkatiman/continuous-improvement/ci.yml?label=CI" alt="CI"></a>
+</p>
+
+<p align="center">
+  <a href="docs/README.zh-CN.md">简体中文</a> |
+  <a href="docs/README.ja.md">日本語</a>
 </p>
 
 ---
@@ -102,7 +111,7 @@ npx continuous-improvement install
 
 That's it. For Claude Code, this installs:
 - Observation hooks (captures every tool call, <50ms, jq optional)
-- `/continuous-improvement` command
+- `/continuous-improvement`, `/discipline`, and `/dashboard` commands
 - Auto-leveling instinct system
 
 ### Expert — full power with MCP server
@@ -112,7 +121,7 @@ npx continuous-improvement install --mode expert
 ```
 
 Everything in beginner plus:
-- **MCP server** with 8 tools (instinct management, import/export, observation viewer)
+- **MCP server** with 10 tools (instinct management, import/export, dashboard, instinct packs)
 - **Session hooks** (auto-load instincts at start, remind to reflect at end)
 - Works with Claude Code, Claude Desktop, and any MCP client
 
@@ -124,6 +133,14 @@ npx continuous-improvement install --mode mcp
 
 Registers the MCP server without hooks — for Cursor, Zed, Windsurf, VS Code, or any editor that supports MCP.
 
+### Load a starter instinct pack
+
+```bash
+npx continuous-improvement install --pack react    # React/Next.js instincts
+npx continuous-improvement install --pack python   # Python best practices
+npx continuous-improvement install --pack go       # Go idioms
+```
+
 ### Install to a specific target
 
 ```bash
@@ -195,14 +212,110 @@ Install:       Hooks start capturing silently. You notice nothing.
 4. **Self-correcting** — user corrections drop confidence by 0.1. Unused instincts decay. Wrong behaviors fade out.
 5. **Project-scoped** — instincts are per-project by default, promoted to global when seen across 2+ projects
 
+### Starter Instinct Packs
+
+Jump-start your instincts with pre-built packs for popular stacks:
+
+```bash
+npx continuous-improvement install --pack react    # 8 React/Next.js instincts
+npx continuous-improvement install --pack python   # 8 Python instincts
+npx continuous-improvement install --pack go       # 8 Go instincts
+```
+
+Or in expert mode: use the `ci_load_pack` tool to load packs at any time.
+
 ### Check what your agent has learned
 
 ```
-/continuous-improvement
+/continuous-improvement    # Reflect, analyze, show status
+/discipline                # Quick reference card of the 7 Laws
+/dashboard                 # Visual instinct health dashboard
+```
+
+---
+
+## GitHub Action: Agent Transcript Linter
+
+Lint your AI agent's behavior in CI/CD. The only GitHub Action that checks if your agent followed disciplined workflows.
+
+```yaml
+- uses: naimkatiman/continuous-improvement@v3
+  with:
+    transcript-path: agent-log.jsonl
+    strict: true  # Fail build on law violations
+```
+
+The linter analyzes tool call patterns and detects:
+- **Law 1 violations** — writes without prior research
+- **Law 3 violations** — too many consecutive edits without verification
+- **Law 4 violations** — code changes without running tests/builds
+- **Law 6 violations** — too many files modified at once
+
+Output includes a discipline score (0-100) and detailed violation report.
+
+```bash
+# Run locally
+node bin/lint-transcript.mjs observations.jsonl
+cat transcript.jsonl | node bin/lint-transcript.mjs --stdin --json
 ```
 
 ---
 
+## Plugin Architecture
+
+continuous-improvement ships as a **plugin** with three layers. Pick what you need:
+
+### Layer 1: Skill Only (any LLM)
+Paste SKILL.md into your system prompt. Your agent follows the 7 Laws. No tools, no hooks, no server.
+
+### Layer 2: Hooks (Claude Code)
+`npx continuous-improvement install` — installs hooks that silently capture every tool call. The instinct system grows automatically. Zero config.
+
+### Layer 3: MCP Server (any MCP client)
+`npx continuous-improvement install --mode expert` — a full MCP server that any editor can connect to.
+
+### Beginner vs Expert
+
+| Feature | Beginner (default) | Expert |
+|---------|-------------------|--------|
+| Observation hooks | Yes | Yes |
+| `/continuous-improvement` command | Yes | Yes |
+| `/discipline` quick reference | Yes | Yes |
+| `/dashboard` visual dashboard | Yes | Yes |
+| Auto-leveling instincts | Yes | Yes |
+| `ci_status` tool | - | Yes |
+| `ci_instincts` tool | - | Yes |
+| `ci_reflect` tool | - | Yes |
+| `ci_reinforce` tool | - | Yes |
+| `ci_create_instinct` tool | - | Yes |
+| `ci_observations` tool | - | Yes |
+| `ci_export` / `ci_import` | - | Yes |
+| `ci_dashboard` tool | - | Yes |
+| `ci_load_pack` tool | - | Yes |
+| Session start/end hooks | - | Yes |
+| MCP server | - | Yes |
+
+**Beginner** is the right choice for 90% of users. It just works — install and forget. The system quietly learns from your sessions.
+
+**Expert** adds the MCP server for programmatic access, manual instinct management, import/export for team sharing, visual dashboard, and instinct packs.
+
+### MCP Tools Reference
+
+| Tool | Description |
+|------|-------------|
+| `ci_status` | Current level, instinct count, observation count |
+| `ci_instincts` | List learned instincts with confidence levels |
+| `ci_reflect` | Generate structured session reflection |
+| `ci_reinforce` | Accept/reject instinct suggestions (expert) |
+| `ci_create_instinct` | Manually create instincts (expert) |
+| `ci_observations` | View raw tool call observations (expert) |
+| `ci_export` | Export instincts as JSON (expert) |
+| `ci_import` | Import instincts from JSON (expert) |
+| `ci_dashboard` | Visual dashboard with confidence distribution (expert) |
+| `ci_load_pack` | Load starter instinct packs (expert) |
+
+---
+
 ## Real-World Examples
 
 See the [`examples/`](examples/) directory for detailed walkthroughs:
@@ -219,22 +332,49 @@ Each example shows the same task done with and without the 7 laws, highlighting
 
 ```
 continuous-improvement/
-├── SKILL.md                    # The 7 Laws + instinct behavior
-├── commands/continuous-improvement.md  # /continuous-improvement command
-├── hooks/observe.sh            # Observation hook (pure bash, <50ms)
-├── bin/install.mjs             # CLI installer
-├── test/                       # 20 tests (node --test)
-├── examples/                   # Real-world before/after scenarios
-├── QUICKSTART.md               # First-use guide
+├── SKILL.md                           # The 7 Laws + instinct behavior
+├── bin/
+│   ├── install.mjs                    # CLI installer (--mode beginner|expert|mcp)
+│   ├── mcp-server.mjs                # MCP server (zero dependencies)
+│   └── lint-transcript.mjs           # Agent transcript linter (GitHub Action)
+├── hooks/
+│   ├── observe.sh                     # Observation hook (pure bash, <50ms)
+│   └── session.sh                     # Session start/end hook (expert mode)
+├── plugins/
+│   ├── beginner.json                  # Plugin manifest: 3 tools
+│   └── expert.json                    # Plugin manifest: 10 tools
+├── commands/
+│   ├── continuous-improvement.md     # /continuous-improvement command
+│   ├── discipline.md                 # /discipline quick reference
+│   └── dashboard.md                  # /dashboard visual display
+├── instinct-packs/
+│   ├── react.json                     # React/Next.js starter instincts
+│   ├── python.json                   # Python starter instincts
+│   └── go.json                       # Go starter instincts
+├── test/                              # 104 tests (node --test)
+├── examples/                          # Real-world before/after scenarios
+├── docs/                              # Translations (zh-CN, ja)
+├── .github/
+│   ├── workflows/ci.yml             # CI pipeline (Node 18/20/22)
+│   └── ISSUE_TEMPLATE/              # Bug report + feature request templates
+├── action.yml                        # GitHub Action definition
+├── llms.txt                          # LLM-friendly project description
+├── CONTRIBUTING.md
+├── CODE_OF_CONDUCT.md
+├── SECURITY.md
+├── QUICKSTART.md
 ├── CHANGELOG.md
 └── package.json
 ```
 
-### What gets installed where (Claude Code)
+### What gets installed where
 
+**Beginner mode** (default):
 ```
 ~/.claude/skills/continuous-improvement/SKILL.md     # The skill
 ~/.claude/commands/continuous-improvement.md          # The command
+~/.claude/commands/discipline.md                     # Quick reference
+~/.claude/commands/dashboard.md                      # Dashboard
 ~/.claude/instincts/
 ├── observe.sh                                       # Hook script
 ├── global/                                          # Global instincts (*.yaml)
@@ -244,6 +384,12 @@ continuous-improvement/
     └── *.yaml                                       # Project instincts
 ```
 
+**Expert mode** adds:
+```
+~/.claude/instincts/session.sh                       # Session hooks
+~/.claude/settings.json                              # + MCP server + session hooks
+```
+
 ---
 
 ## Uninstall
@@ -252,7 +398,7 @@ continuous-improvement/
 npx continuous-improvement install --uninstall
 ```
 
-Removes the skill, hooks, and command. Your learned instincts in `~/.claude/instincts/` are preserved — delete that directory manually if you want a clean slate.
+Removes the skill, hooks, and commands. Your learned instincts in `~/.claude/instincts/` are preserved — delete that directory manually if you want a clean slate.
 
 ---
 
@@ -260,8 +406,11 @@ Removes the skill, hooks, and command. Your learned instincts in `~/.claude/inst
 
 | Tool | Support |
 |------|---------|
-| **Claude Code** | Full — skill + hooks + auto-leveling instincts |
-| **Cursor** | Skill only (paste SKILL.md into rules) |
+| **Claude Code** | Full — skill + hooks + MCP server + auto-leveling instincts |
+| **Claude Desktop** | MCP server (expert/mcp mode) |
+| **Cursor** | MCP server (mcp mode) or skill only (paste SKILL.md into rules) |
+| **Zed / Windsurf** | MCP server (mcp mode) |
+| **VS Code** | MCP server (mcp mode) with Copilot MCP support |
 | **Codex** | Skill only |
 | **Gemini CLI** | Skill only |
 | **OpenClaw** | Skill only |
@@ -286,33 +435,62 @@ If your agent says any of these, it's skipping a law:
 ### Phase 1: Foundation -- DONE
 
 - [x] Published to public npm (`npx continuous-improvement install` works)
-- [x] 20-test suite (installer, hook, SKILL.md validation)
+- [x] 104-test suite (installer, hook, MCP server, linter, packs, community files)
 - [x] Before/after examples in README + `examples/` directory
 - [x] Gemini CLI support
 - [x] Platform badges and improved npm metadata
 - [ ] **Submit to [awesome-agent-skills](https://github.com/VoltAgent/awesome-agent-skills)** (14K stars)
 
-### Phase 2: Content & Proof (In Progress)
+### Phase 2: Plugin Architecture -- DONE
+
+- [x] **MCP server** — 10 tools (beginner: 3, expert: 7 more) with zero dependencies
+- [x] **Beginner / Expert separation** — simple defaults, power when you need it
+- [x] **Plugin manifests** — `plugins/beginner.json` and `plugins/expert.json`
+- [x] **Session hooks** — auto-load instincts at session start, remind to reflect at end
+- [x] **`--mode` flag** — `beginner` | `expert` | `mcp` installation modes
+- [x] **Import/export** — share instincts as JSON between team members
+- [x] **Multi-editor MCP support** — Claude Desktop, Cursor, Zed, Windsurf, VS Code
+
+### Phase 2.5: Visibility & Ecosystem -- DONE
+
+- [x] **GitHub Action** — lint agent transcripts for law compliance (`action.yml`)
+- [x] **Starter instinct packs** — React, Python, Go (pre-built instincts)
+- [x] **`/discipline` command** — quick reference card of the 7 Laws
+- [x] **`/dashboard` command** — visual instinct health dashboard
+- [x] **Community files** — CONTRIBUTING.md, CODE_OF_CONDUCT.md, SECURITY.md
+- [x] **llms.txt** — LLM-friendly project description for discoverability
+- [x] **CI pipeline** — GitHub Actions testing on Node 18/20/22
+- [x] **Issue templates** — bug report + feature request
+- [x] **Translations** — Chinese (简体中文) and Japanese (日本語)
+
+### Phase 3: Content & Proof
 
 - [ ] **2-min demo video** — side-by-side agent with/without discipline. Post to X + YouTube.
 - [ ] **"Why your AI agent keeps lying about being done"** — X thread / blog post
 - [ ] **"Law of the Week" X series** — 7 weeks of content breaking down each law
 
-### Phase 3: Ecosystem Integration
+### Phase 4: Ecosystem Growth
 
-- [ ] **MCP server** — expose instinct status + law compliance as MCP tools
-- [ ] **GitHub Action** — lint agent transcripts for law compliance
 - [ ] **VS Code extension** — sidebar showing instinct confidence levels
-- [ ] **Aider / Windsurf / Zed** support
+- [ ] **More instinct packs** — TypeScript, Rust, Java, Django, Laravel
+- [ ] **Instinct marketplace** — share learned instincts across teams
 
-### Phase 4: Community
+### Phase 5: Community
 
-- [ ] **Instinct marketplace** — share learned instincts across teams
-- [ ] **Quick-start instinct packs** — pre-built instincts for React, Python, Go, etc.
 - [ ] **Conference talk on Mulahazah** — the auto-leveling system is genuinely novel
+- [ ] **Leaderboard / badges** — "100 sessions" achievement system
+- [ ] **Custom domain** — landing page with interactive demo
 
 ---
 
+## Contributing
+
+See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines. Issues and PRs welcome.
+
+## Security
+
+See [SECURITY.md](SECURITY.md) for the security policy and how to report vulnerabilities.
+
 ## License
 
 MIT

package/action.yml

@@ -0,0 +1,33 @@
+name: 'AI Agent Discipline Linter'
+description: 'Lint AI agent transcripts for compliance with the 7 Laws of AI Agent Discipline'
+author: 'naimkatiman'
+
+branding:
+  icon: 'check-circle'
+  color: 'green'
+
+inputs:
+  transcript-path:
+    description: 'Path to the agent transcript file (JSONL format)'
+    required: false
+    default: ''
+  observations-path:
+    description: 'Path to observations.jsonl from a CI session'
+    required: false
+    default: ''
+  strict:
+    description: 'Fail the check if any law violations are detected'
+    required: false
+    default: 'false'
+
+outputs:
+  violations:
+    description: 'Number of law violations found'
+  score:
+    description: 'Discipline score (0-100)'
+  report:
+    description: 'Markdown-formatted report'
+
+runs:
+  using: 'node20'
+  main: 'bin/lint-transcript.mjs'

package/bin/install.mjs

@@ -23,10 +23,12 @@ import {
   writeFileSync,
   rmSync,
   chmodSync,
+  readdirSync,
 } from "node:fs";
 import { join, dirname } from "node:path";
 import { homedir } from "node:os";
 import { fileURLToPath } from "node:url";
+import { execSync } from "node:child_process";
 
 const __filename = fileURLToPath(import.meta.url);
 const __dirname = dirname(__filename);
@@ -133,14 +135,17 @@ function setupMulahazah() {
     }
   }
 
-  // 4. Copy /continuous-improvement command
+  // 4. Copy all commands
   const commandsDir = join(home, ".claude", "commands");
   mkdirSync(commandsDir, { recursive: true });
-  const cmdSrc = join(REPO_ROOT, "commands", "continuous-improvement.md");
-  const cmdDest = join(commandsDir, "continuous-improvement.md");
-  if (existsSync(cmdSrc)) {
-    copyFileSync(cmdSrc, cmdDest);
-    console.log(`  ✓ /continuous-improvement command → ${cmdDest}`);
+  const commandFiles = ["continuous-improvement.md", "discipline.md", "dashboard.md"];
+  for (const cmdFile of commandFiles) {
+    const cmdSrc = join(REPO_ROOT, "commands", cmdFile);
+    const cmdDest = join(commandsDir, cmdFile);
+    if (existsSync(cmdSrc)) {
+      copyFileSync(cmdSrc, cmdDest);
+      console.log(`  ✓ /${cmdFile.replace(".md", "")} command → ${cmdDest}`);
+    }
   }
 
   // 5. Patch ~/.claude/settings.json with hooks
@@ -301,14 +306,16 @@ function uninstallAll() {
     }
   }
 
-  // 2. Remove /continuous-improvement command
-  const cmdFile = join(home, ".claude", "commands", "continuous-improvement.md");
-  if (existsSync(cmdFile)) {
-    try {
-      rmSync(cmdFile);
-      console.log(`  ✓ Removed /continuous-improvement command`);
-    } catch (err) {
-      console.error(`  ✗ Command file: ${err.message}`);
+  // 2. Remove all commands
+  for (const cmdName of ["continuous-improvement.md", "discipline.md", "dashboard.md"]) {
+    const cmdFile = join(home, ".claude", "commands", cmdName);
+    if (existsSync(cmdFile)) {
+      try {
+        rmSync(cmdFile);
+        console.log(`  ✓ Removed /${cmdName.replace(".md", "")} command`);
+      } catch (err) {
+        console.error(`  ✗ ${cmdName}: ${err.message}`);
+      }
     }
   }
 
@@ -442,7 +449,7 @@ if (args.includes("--uninstall")) {
 }
 
 console.log(`
-continuous-improvement v3.0 (mode: ${INSTALL_MODE})
+continuous-improvement v3.1 (mode: ${INSTALL_MODE})
 Research → Plan → Execute → Verify → Reflect → Learn → Iterate
 `);
 
@@ -485,6 +492,58 @@ const modeInfo = {
   mcp: "MCP server registered. Connect from any MCP-compatible editor.",
 };
 
+// Handle --pack flag
+const packIdx = _args.indexOf("--pack");
+if (packIdx !== -1 && _args[packIdx + 1]) {
+  const packName = _args[packIdx + 1];
+  const packPath = join(REPO_ROOT, "instinct-packs", `${packName}.json`);
+  if (existsSync(packPath)) {
+    const project = getProjectHashSync();
+    const targetDir = join(homedir(), ".claude", "instincts", project.hash);
+    mkdirSync(targetDir, { recursive: true });
+
+    const instincts = JSON.parse(readFileSync(packPath, "utf8"));
+    let loaded = 0;
+    for (const inst of instincts) {
+      const instPath = join(targetDir, `${inst.id}.yaml`);
+      if (!existsSync(instPath)) {
+        const yaml = [
+          `id: ${inst.id}`,
+          `trigger: "${inst.trigger}"`,
+          `confidence: ${inst.confidence}`,
+          `domain: ${inst.domain || "workflow"}`,
+          `source: pack-${packName}`,
+          `scope: project`,
+          `project_id: ${project.hash}`,
+          `created: "${new Date().toISOString().split("T")[0]}"`,
+          `last_seen: "${new Date().toISOString().split("T")[0]}"`,
+          `observation_count: 0`,
+          "---",
+          inst.body,
+        ].join("\n");
+        writeFileSync(instPath, yaml + "\n");
+        loaded++;
+      }
+    }
+    console.log(`  ✓ Loaded ${loaded}/${instincts.length} instincts from ${packName} pack`);
+  } else {
+    const available = readdirSync(join(REPO_ROOT, "instinct-packs"))
+      .filter((f) => f.endsWith(".json"))
+      .map((f) => f.replace(".json", ""));
+    console.error(`  ✗ Unknown pack: ${packName}. Available: ${available.join(", ")}`);
+  }
+}
+
+function getProjectHashSync() {
+  try {
+    const root = execSync("git rev-parse --show-toplevel 2>/dev/null", { encoding: "utf8" }).trim();
+    const hash = execSync(`printf '%s' "${root}" | sha256sum | cut -c1-12`, { encoding: "utf8", shell: "/bin/bash" }).trim();
+    return { root, hash };
+  } catch {
+    return { root: "global", hash: "global" };
+  }
+}
+
 console.log(`
 ${installed > 0 ? "Done." : "Failed."} Installed to ${installed}/${targets.length} target(s).
 ${hasClaude ? `\n${modeInfo[INSTALL_MODE] || modeInfo.beginner}` : ""}
@@ -492,6 +551,8 @@ Next steps:
   1. Start a new Claude Code session
   2. Say: "Use the continuous-improvement framework to [your task]"
   3. After your first task, run: /continuous-improvement
-${INSTALL_MODE === "expert" ? "\nMCP tools available: ci_status, ci_instincts, ci_reflect, ci_reinforce,\n  ci_create_instinct, ci_observations, ci_export, ci_import" : ""}
+  4. Try: /discipline for quick reference, /dashboard for instinct health
+${INSTALL_MODE === "expert" ? "\nMCP tools available: ci_status, ci_instincts, ci_reflect, ci_reinforce,\n  ci_create_instinct, ci_observations, ci_export, ci_import, ci_dashboard, ci_load_pack" : ""}
 ${INSTALL_MODE === "mcp" ? "\nMCP tools available: ci_status, ci_instincts, ci_reflect" : ""}
+Available instinct packs: npx continuous-improvement install --pack react|python|go
 `);