@techwavedev/agi-agent-kit 1.4.2 → 1.5.2

package/CHANGELOG.md

@@ -5,6 +5,52 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [1.5.2] - 2026-02-21
+
+### Added
+
+- **Portuguese Localization** — Added `README.pt-BR.md` to support Brazilian Portuguese speakers.
+- **UI UX Pro Max Prompts** — Expanded the `ui-ux-pro-max` skill with proven structural and stylistic prompt examples (SaaS, Educational, Pet Grooming, AI Chatbot).
+
+## [1.5.1] - 2026-02-21
+
+### Changed
+
+- **UI UX Pro Max v2.0** — Upgraded design intelligence skill with major enhancements:
+  - **Design System Generator** — AI-powered reasoning engine that analyzes project requirements and generates complete, tailored design systems (pattern, style, colors, typography, effects, anti-patterns)
+  - **100 Industry-Specific Reasoning Rules** — Automated style/color/typography selection based on product type and industry
+  - **Expanded Data** — 67 UI styles (was 50), 96 color palettes (was 21), 57 font pairings (was 50), 25 chart types (was 20), 13 tech stacks (was 9), 99 UX guidelines
+  - **Persist Design System** — Master + Overrides pattern for hierarchical retrieval across sessions (`--persist` flag)
+  - **Qdrant Memory Integration** — Design decisions automatically stored and retrieved for project continuity
+  - **SKILLS_CATALOG.md** — Updated `ui-ux-pro-max` description to reflect v2.0 capabilities
+
+## [1.5.0] - 2026-02-20
+
+### Added
+
+- **Smart Init Wizard** — Complete overhaul of `npx init` with a guided, step-by-step setup experience:
+  - **Existing install detection** — Reads `.agi-version` stamp; offers `Update` (preserve `.env`), `Reinstall` (full overwrite), or `Cancel`. Shows installed vs incoming version with downgrade warning.
+  - **Install scope prompt** — Choose between project-local (current dir) or global (`~/.agent` + platform symlinks) with a compatibility table and pitfalls disclaimer.
+  - **Smart backup** — Scans files that would be overwritten before touching anything. For global installs, also detects real platform dirs that would be replaced by symlinks. Saves timestamped backup.
+  - **Custom domain pack selection** — New `4. custom` option shows all 15 skill domains with ■ professional and ■ community skill counts, supports comma/range multi-select (`1,3,7-9`, `all`).
+  - **Service-aware memory setup** — Detects Ollama, Docker, and Qdrant before asking. If a service is missing, asks whether it's not installed yet (shows install link) or running on a custom URL (prompts for URL + API key, verifies connectivity).
+  - **Agent Teams prompt** — Explicit opt-in to `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS` with explanation and safe merge into `.claude/settings.json`.
+  - **Uninstall script** — Global installs generate `~/.agent/uninstall-agi.sh`, a ready-to-run script that removes all symlinks and the install directory.
+  - **Version stamp** — Writes `.agi-version` after every install/update.
+  - **Configurable final summary** — Shows exactly what was configured (memory, Agent Teams, MCP) and what manual steps remain (plugins, MCP servers).
+
+### Changed
+
+- **`verifyMemorySetup()`** — Now returns `true`/`false` so the final message correctly shows "Memory is READY" vs "start services" only when relevant.
+- **`writeEnvFile()`** — Now writes the actual Qdrant URL, Ollama URL, and API key entered during setup instead of always defaulting to localhost.
+- **`copySkills()`** — Refactored to share a helper. Custom pack installs core first then resolves each selected domain from both `knowledge/` and `extended/` tiers.
+- **Platform setup** (`runPlatformSetup`) — Now runs _after_ the new `promptPlatformFeatures` step so user intent is captured before platform scripts apply settings.
+
+### Fixed
+
+- **Next steps message** — No longer suggests starting services when they were already verified during install.
+- **`memoryVerified` bug** — Was referenced but never assigned; `verifyMemorySetup()` return value is now captured.
+
 ## [1.4.2]
 
 - **Fixed README links** — ./skills/SKILLS_CATALOG.md updated to ./templates/skills/SKILLS_CATALOG.md (closes #15)

package/README.md

@@ -1,5 +1,7 @@
 # 🚀 AGI Agent Kit
 
+🌐 _[Português (BR)](./README.pt-BR.md)_ | English
+
 > **Stop hallucinating. Start executing.**
 
 [![npm version](https://img.shields.io/npm/v/@techwavedev/agi-agent-kit.svg)](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)
@@ -18,7 +20,7 @@
 
 **AGI Agent Kit** is the enterprise-grade scaffolding that turns any AI coding assistant into a **deterministic production machine**. While LLMs are probabilistic (90% accuracy per step = 59% over 5 steps), this framework forces them through a **3-Layer Architecture** — Intent → Orchestration → Execution — where business logic lives in tested scripts, not hallucinated code.
 
-### Why this exists
+## Why this exists
 
 Most AI coding setups give you a prompt and hope for the best. AGI Agent Kit gives you:
 
@@ -42,29 +44,34 @@ Scaffold a new agent workspace in seconds:
 
 ```bash
 npx @techwavedev/agi-agent-kit init
-```
-
-You'll be prompted to choose a pack:
-
-- **core** — 4 essential skills (webcrawler, pdf-reader, qdrant-memory, documentation)
-- **medium** — Core + 89 specialized skills in 16 categories + `.agent/` structure (API, Security, Design, Architecture)
-- **full** — Complete suite: Medium + 785 community skills from [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills) (878 total)
 
-After installation, run the **one-shot setup wizard** to auto-configure your environment:
-
-```bash
-python3 skills/plugin-discovery/scripts/platform_setup.py --project-dir .
+# Or install globally to ~/.agent to share skills across projects
+npx @techwavedev/agi-agent-kit init --global
 ```
 
-This detects your platform, scans the project stack, and configures everything with a single confirmation.
+You'll be guided through an interactive wizard:
 
-Then **boot the memory system** for automatic token savings:
+1. **Existing install check** — detects a prior install and offers Update / Reinstall / Cancel
+2. **Install scope** — project-local (current dir) or global (`~/.agent` shared across projects)
+3. **Smart backup** — scans files at risk and creates a timestamped backup before touching anything
+4. **Pack selection** — choose skills to install:
+   - **core** — 4 essential skills (webcrawler, pdf-reader, qdrant-memory, documentation)
+   - **medium** — Core + 89 professional skills in 16 categories + `.agent/` structure
+   - **full** — Everything: Medium + 785 community skills (878 total)
+   - **custom** — Core + you pick specific domains (AI Agents, DevOps, Security, Frontend, etc.)
+5. **Memory setup** — detects Ollama/Docker/Qdrant; if missing, asks whether to install locally or use a custom URL (supports Qdrant Cloud, remote servers)
+6. **Agent Teams** — opt-in to parallel multi-agent execution (writes `.claude/settings.json`)
+7. **Summary** — shows exactly what was configured vs what needs manual action
+
+After installation the wizard shows your next steps, including:
 
 ```bash
+# Boot the memory system (verifies Qdrant + Ollama, auto-fixes issues)
 python3 execution/session_boot.py --auto-fix
-```
 
-This checks Qdrant, Ollama, embedding models, and collections — auto-fixing any issues.
+# Run the platform setup wizard (auto-configures your AI platform)
+python3 skills/plugin-discovery/scripts/platform_setup.py --project-dir .
+```
 
 ---
 
@@ -112,7 +119,7 @@ The agi framework adopts all best patterns from [obra/superpowers](https://githu
 
 The framework supports two orchestration modes. Here are **real test results** from `execution/benchmark_modes.py` running on local infrastructure (Qdrant + Ollama `nomic-embed-text`, zero cloud API calls):
 
-```
+```text
 MODE A: SUBAGENTS — Independent, fire-and-forget
   📤 Explore Auth Patterns    → ✅ stored in cache + memory (127ms)
   📤 Query Performance        → ❌ FAILED (timeout — fault tolerant)
@@ -213,7 +220,7 @@ python3 skills/plugin-discovery/scripts/platform_setup.py --project-dir . --dry-
 
 ## 📦 What You Get
 
-```
+```text
 your-project/
 ├── AGENTS.md              # Master instruction file
 ├── GEMINI.md → AGENTS.md  # Platform symlinks
@@ -244,7 +251,7 @@ your-project/
 
 The system operates on three layers:
 
-```
+```text
 ┌─────────────────────────────────────────────────────────┐
 │  Layer 1: DIRECTIVES (Intent)                           │
 │  └─ SOPs written in Markdown (directives/)              │
@@ -341,6 +348,8 @@ If you need to reinstall or update dependencies:
 
 ```bash
 npx @techwavedev/agi-agent-kit init --pack=full
+# To install globally instead of per-project:
+npx @techwavedev/agi-agent-kit init --pack=full --global
 ```
 
 ### Auto-detect platform and configure environment