agents-united 1.1.6 → 1.2.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (5) hide show
  1. package/README.md +200 -71
  2. package/dist/cli.js +21 -21
  3. package/dist/cli.js.map +1 -1
  4. package/package.json +1 -1
  5. package/registry/bundles.json +10 -6
package/README.md CHANGED
@@ -2,7 +2,7 @@
2
2
 
3
3
  # Agents United
4
4
 
5
- [![npm version](https://img.shields.io/badge/npm-v1.0.0-blue.svg)](https://www.npmjs.com/package/agents-united)
5
+ [![npm version](https://img.shields.io/npm/v/agents-united.svg?color=blue)](https://www.npmjs.com/package/agents-united)
6
6
  [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
7
7
  [![Tests: Vitest](https://img.shields.io/badge/tests-128%2B%20passing-brightgreen.svg)](https://vitest.dev/)
8
8
  [![TypeScript: Strict](https://img.shields.io/badge/TypeScript-Strict-blue.svg)](https://www.typescriptlang.org/)
@@ -15,31 +15,41 @@ Curated teams of orchestrators, sub-agents, skills, and workflows — installed
15
15
 
16
16
  ## 🌟 Key Features
17
17
 
18
- - **ðŸŠķ Small Footprint by Default, On-Demand Growth**: Every department installs as a lean **Essentials bundle** by default. When a task exceeds the installed capability, the Lead Orchestrator automatically detects the gap, names the addon bundle that covers it, and — in CLI-enabled environments — installs it directly into the scope of the active main domain (e.g. `agents add seo-content-marketing`) without any manual discovery needed.
19
- - **Department Domain Hierarchy**: Discover and install curated teams organized into 8 functional departments (Software Engineering, System Architecture, Product Design, Growth & Marketing, Security, Research, Business Strategy, and Universal Suite).
20
- - **18 Curated Bundles**: 8 Essentials foundational base bundles + 10 specialized sub-team addons (including AI/ML Engineering, Mobile, DevOps, SRE, SEO, Paid PPC, PLG, and Lifecycle Email).
21
- - **Cross-Bundle Dynamic Recommendation Protocol**: All 7 Lead Orchestrators possess real-time knowledge of specialized sub-domain bundles and recommend precise `agents add <bundle>` or `agents add domain:<dept>` commands when complex domain tasks are requested.
22
- - **Modern Cloud Platform & AI Ecosystem**: First-class runbooks for Modal.com, Replicate, RunPod, Ollama, vLLM, LangChain, LlamaIndex, Qdrant, Vercel, Lovable/v0, Supabase, Turso, and Azure Bicep.
23
- - **Scoped AI Safety Policies**: Built-in zero secret exposure, GPU cost ceilings (scale-to-zero timeouts), and training data PII scrubbing.
24
- - **Two-Stage Progressive TUI**: Interactive `@clack/prompts` drill-down menu for effortless browsing without command-line parameter memorization.
25
- - **Folder Tree Visualizations**: Human-friendly terminal catalog view (`agents list`) displaying orchestrators, sub-agents, skills, workflows, and bundle inheritance.
26
- - **Multi-Host Target Adapters**: Simultaneously equip `.agents/`, `.gemini/`, `.claude/`, and `.cursor/` runtimes with automatic workspace detection.
27
- - **Flexible Scopes & Installation Modes**: Support for **Project Scope** (Git-tracked team lockfiles) and **Global Scope** (`~/.agents/`), with zero-duplication **Symlink Mode** or isolated **Copy Mode**.
28
- - **Whole-Department Installation**: One-click installation of entire department suites (`domain:engineering`, `domain:marketing`, `domain:architecture`) with interactive safety confirmation prompts.
29
- - **Fast Search & Discovery**: Powerful query engine (`agents find`) with category/type filters and interactive one-click installation.
30
- - **Strict Git Guardrails & Zero-Trust Safety**: Built-in protection against commits to protected branches (`main`/`master`), accidental force pushes, and secret exposure.
31
- - **Deterministic Health Doctor**: Automatic validation of frontmatter schema compliance, lifecycle hooks, and lockfile synchronization (`agents doctor`).
18
+ - **ðŸŠķ Essentials-First & On-Demand Growth**: Every department installs as a lean **Essentials bundle** by default. When a task requires specialized capabilities, the Lead Orchestrator automatically detects the gap, recommends the exact addon, and can auto-install it directly into your project scope.
19
+ - **🌐 One Library, Every Assistant**: Author in `.agents/` as your single source of truth. Agents United automatically projects and translates compatible configurations to **Google Antigravity 2.0 / Gemini**, **Anthropic Claude Code**, **Cursor**, **Cline**, **OpenCode**, and **Codex / AGENTS.md**.
20
+ - **🚀 Cline Native Activation**: Launch specialized teams into Cline CLI sessions with `agents start`, featuring compound role projections, skill toolkits, coordinator rules, and declarative team manifests.
21
+ - **🏛ïļ 8 Department Domains & 22 Bundles**: Complete coverage across Software Engineering, System Architecture, Product Design, Growth & Marketing, Security, Deep Research, Business Strategy, and Universal Meta-Skills.
22
+ - **ðŸĪ– 45 Specialized Agents & 90+ Skills**: 7 Lead Orchestrators coordinating 38 domain sub-agents, backed by 90 production-grade runbooks and 63 deterministic workflows.
23
+ - **⚡ Modern Cloud & AI Tooling**: First-class runbooks for Modal.com, Replicate, RunPod, local LLMs/vLLM, LangChain, LlamaIndex, Qdrant, Vercel, Supabase, Turso, and Azure Bicep.
24
+ - **ðŸĒ Organization Bundles (Tier 2 / Experimental)**: Cross-functional multi-disciplinary teams *(Currently under development)*.
25
+ - **ðŸ›Ąïļ Built-in Zero-Trust Safety & Git Guardrails**: Hard-coded branch protection (`main`/`master`), zero force-push policy, secret redaction, serverless GPU cost ceilings, and PII scrubbing.
26
+ - **🧙 Interactive 2-Stage TUI Wizard**: Guided terminal interface powered by `@clack/prompts`, visual Unicode catalog tree (`agents list`), and fast search (`agents find`).
27
+ - **ðŸ“Ķ Flexible Scopes & Installation Modes**: Install to **Project Scope** (Git-tracked team lockfiles) or **Global Scope** (`~/.agents/`), using instant **Symlink Mode** or standalone **Copy Mode**.
28
+ - **ðŸĐš Deterministic Health Doctor**: Comprehensive diagnostic suite (`agents doctor`) validating frontmatter schemas, declarative lifecycle hooks, host capabilities, and lockfile synchronization.
32
29
 
33
30
  ---
34
31
 
35
32
  ## 🚀 Quickstart
36
33
 
37
- ### Run with `npx` (Zero Installation)
34
+ ### 1. Run with `npx` (Zero Installation)
35
+ Try immediately without installing anything:
38
36
  ```bash
39
37
  npx agents-united add
40
38
  ```
41
39
 
42
- ### Global Installation
40
+ ### 2. Project-Level Installation (Recommended for Teams)
41
+ Install into your project repository to track agents and share lockfiles with your team:
42
+ ```bash
43
+ # Install as a dev dependency
44
+ npm install -D agents-united
45
+ # or: pnpm add -D agents-united | yarn add -D agents-united
46
+
47
+ # Launch interactive setup for your repository (installs into ./.agents/)
48
+ npx agents add
49
+ ```
50
+
51
+ ### 3. Global Installation
52
+ Install globally to use the `agents` CLI anywhere on your machine:
43
53
  ```bash
44
54
  npm install -g agents-united
45
55
  ```
@@ -99,17 +109,72 @@ This model gives you:
99
109
 
100
110
  Agents United provides **22 curated bundles**, **45 specialized agents** (7 Lead Orchestrators + 38 Sub-Agents), **90 skills**, and **63 workflows** across 8 department domains:
101
111
 
102
- | Department Domain | Bundles | Description | Orchestrator & Key Roles |
103
- |---|---|---|---|
104
- | **🌐 Universal Autonomous Department** | `universal-skills` ⭐ *(Recommended Baseline)*<br>`full` (Universal Suite) | Domain-agnostic meta-skills (Socratic grilling, PRD generation, ADRs, session handoff) and complete enterprise suite containing all 22 bundles and 45 agents. | All 7 Lead Orchestrators + 38 Sub-Agents |
105
- | **🛠ïļ Software Engineering & Delivery** | `software-engineering` (Essentials Base)<br>`ai-ml-engineering`<br>`mobile-development`<br>`frontend-engineering`<br>`backend-distributed-systems`<br>`qa-automation`<br>`devops-engineering` | Full lifecycle software engineering, AI/ML platform engineering & serverless GPU orchestration, mobile (iOS/Android/Cross-Platform), modern web frontend, microservices, Playwright E2E automation, and CI/CD delivery pipelines. | `orchestrator-engineering`<br>`subagent-backend-architect`<br>`subagent-frontend-architect`<br>`subagent-code-reviewer`<br>`subagent-repo-index`<br>`subagent-ml-platform-engineer`<br>`subagent-ai-model-architect`<br>`subagent-ios-architect`<br>`subagent-android-architect`<br>`subagent-cross-platform-specialist`<br>`subagent-accessibility-lead`<br>`subagent-distributed-systems-architect`<br>`subagent-data-engineer`<br>`subagent-qa-automation-lead`<br>`subagent-e2e-tester`<br>`subagent-devops-engineer` |
106
- | **🏛ïļ System Architecture & SRE** | `system-architecture` (Essentials Base)<br>`sysops-sre` | High-level system design, distributed data models, ADR governance, 99.999% uptime, Prometheus telemetry, incident triage, and disaster recovery. | `orchestrator-system-architecture`<br>`subagent-system-architect`<br>`subagent-backend-architect`<br>`subagent-sysops-sre-lead` |
107
- | **ðŸŽĻ Product Design & UI/UX** | `product-design` (Essentials Base)<br>`design-systems-ops`<br>`design-research-testing` | User research, wireframing, design systems governance, token management, sprint handoffs, micro-interactions, and AI prototype refactoring. | `orchestrator-design`<br>`subagent-ui-designer`<br>`subagent-ux-strategist`<br>`subagent-interaction-designer`<br>`subagent-design-systems-architect`<br>`subagent-design-ops-lead`<br>`subagent-design-researcher`<br>`subagent-designer-toolkit-expert`<br>`subagent-prototype-tester` |
108
- | **📈 Growth & Marketing Operations** | `growth-marketing` (Essentials Base)<br>`seo-content-marketing`<br>`performance-paid-acquisition`<br>`product-led-growth`<br>`lifecycle-email-marketing` | Growth strategy, visual creative design, programmatic SEO, multi-channel paid acquisition (Google/Meta/LinkedIn), onboarding funnel CRO, viral loops, and behavioral email nurture sequences. | `orchestrator-marketing`<br>`subagent-marketing-growth-strategist`<br>`subagent-marketing-content-strategist`<br>`subagent-marketing-conversion-specialist`<br>`subagent-marketing-campaign-specialist`<br>`subagent-marketing-creative-designer`<br>`subagent-seo-specialist`<br>`subagent-paid-acquisition-specialist`<br>`subagent-plg-strategist`<br>`subagent-lifecycle-email-specialist` |
109
- | **🔒 Security Operations** | `security-operations` (Essentials Base) | Application security audits, STRIDE threat modeling, dependency vulnerability scanning, secret redaction, and infrastructure hardening. | `orchestrator-security`<br>`subagent-security-engineer` |
110
- | **🔎 Deep Technical Research** | `deep-research` (Essentials Base) | Technical literature review, deep research synthesis, Socratic mentoring, and repository indexing. | `orchestrator-research`<br>`subagent-deep-research`<br>`subagent-socratic-mentor`<br>`subagent-repo-index` |
111
- | **💞 Business Strategy & Economics** | `business-strategy` (Essentials Base) | Market sizing, unit economics, SaaS monetization models, and executive technical specification panels. | `orchestrator-business`<br>`subagent-business-panel-experts` |
112
- | **ðŸĒ Organization Bundles (Tier 2 / Experimental)** | `digital-agency` 🚧 *(Under Construction)* | Cross-functional digital product agency combining web dev, mobile, UI/UX, SEO, and paid growth. Requires MCP servers (`firecrawl`, `github`) and packages. Supports Dual Execution Modes (Operational vs Brainstorming). | `orchestrator-engineering`<br>`subagent-backend-architect`<br>`subagent-frontend-architect`<br>`subagent-ui-designer`<br>`subagent-marketing-growth-strategist`<br>`subagent-seo-specialist` |
112
+ | Department Domain | Essentials Base | Specialized Addon Bundles | Lead Orchestrator |
113
+ | :--- | :--- | :--- | :--- |
114
+ | **🌐 Universal Department** | `universal-skills` ⭐ | `full` (Universal Suite) | Multi-Agent Coordination |
115
+ | **🛠ïļ Software Engineering & Delivery** | `software-engineering` | `ai-ml-engineering`, `mobile-development`, `frontend-engineering`, `backend-distributed-systems`, `qa-automation`, `devops-engineering` | `orchestrator-engineering` |
116
+ | **🏛ïļ System Architecture & SRE** | `system-architecture` | `sysops-sre` | `orchestrator-system-architecture` |
117
+ | **ðŸŽĻ Product Design & UI/UX** | `product-design` | `design-systems-ops`, `design-research-testing` | `orchestrator-design` |
118
+ | **📈 Growth & Marketing Operations** | `growth-marketing` | `seo-content-marketing`, `performance-paid-acquisition`, `product-led-growth`, `lifecycle-email-marketing` | `orchestrator-marketing` |
119
+ | **🔒 Security Operations** | `security-operations` | — | `orchestrator-security` |
120
+ | **🔎 Deep Technical Research** | `deep-research` | — | `orchestrator-research` |
121
+ | **💞 Business Strategy & Economics** | `business-strategy` | — | `orchestrator-business` |
122
+ | **ðŸĒ Organization Bundles (Experimental)** | `digital-agency` 🚧 | *(Currently under development)* | `orchestrator-digital-agency-director` |
123
+
124
+ <details>
125
+ <summary><strong>🔍 Click to expand full Department & Agent Roster Breakdown</strong></summary>
126
+
127
+ <br>
128
+
129
+ #### 🌐 Universal Autonomous Department
130
+ - **Baseline**: `universal-skills` ⭐ *(Recommended Baseline — Socratic grilling, PRD generation, ADRs, session handoff)*
131
+ - **Universal Suite**: `full` *(Complete enterprise suite containing all 22 bundles, 45 agents, 90 skills, 63 workflows)*
132
+
133
+ #### 🛠ïļ Software Engineering & Delivery
134
+ - **Lead Orchestrator**: `orchestrator-engineering`
135
+ - **Essentials Base (`software-engineering`)**: `subagent-backend-architect`, `subagent-frontend-architect`, `subagent-code-reviewer`, `subagent-repo-index`
136
+ - **AI/ML Engineering (`ai-ml-engineering`)**: `subagent-ml-platform-engineer`, `subagent-ai-model-architect`
137
+ - **Mobile Development (`mobile-development`)**: `subagent-ios-architect`, `subagent-android-architect`, `subagent-cross-platform-specialist`
138
+ - **Frontend Engineering (`frontend-engineering`)**: `subagent-frontend-architect`, `subagent-accessibility-lead`
139
+ - **Backend & Distributed Systems (`backend-distributed-systems`)**: `subagent-distributed-systems-architect`, `subagent-data-engineer`
140
+ - **QA Automation (`qa-automation`)**: `subagent-qa-automation-lead`, `subagent-e2e-tester`
141
+ - **DevOps Engineering (`devops-engineering`)**: `subagent-devops-engineer`
142
+
143
+ #### 🏛ïļ System Architecture & SRE
144
+ - **Lead Orchestrator**: `orchestrator-system-architecture`
145
+ - **Essentials Base (`system-architecture`)**: `subagent-system-architect`, `subagent-backend-architect`
146
+ - **SysOps & SRE (`sysops-sre`)**: `subagent-sysops-sre-lead`
147
+
148
+ #### ðŸŽĻ Product Design & UI/UX
149
+ - **Lead Orchestrator**: `orchestrator-design`
150
+ - **Essentials Base (`product-design`)**: `subagent-ui-designer`, `subagent-ux-strategist`, `subagent-interaction-designer`, `subagent-design-systems-architect`, `subagent-design-researcher`, `subagent-design-ops-lead`, `subagent-designer-toolkit-expert`, `subagent-prototype-tester`
151
+ - **Design Systems Ops (`design-systems-ops`)**: `subagent-design-systems-architect`, `subagent-design-ops-lead`
152
+ - **Design Research & Testing (`design-research-testing`)**: `subagent-design-researcher`, `subagent-prototype-tester`
153
+
154
+ #### 📈 Growth & Marketing Operations
155
+ - **Lead Orchestrator**: `orchestrator-marketing`
156
+ - **Essentials Base (`growth-marketing`)**: `subagent-marketing-growth-strategist`, `subagent-marketing-content-strategist`, `subagent-marketing-conversion-specialist`, `subagent-marketing-campaign-specialist`, `subagent-marketing-creative-designer`
157
+ - **SEO & Content Marketing (`seo-content-marketing`)**: `subagent-seo-specialist`
158
+ - **Performance & Paid Acquisition (`performance-paid-acquisition`)**: `subagent-paid-acquisition-specialist`
159
+ - **Product-Led Growth (`product-led-growth`)**: `subagent-plg-strategist`
160
+ - **Lifecycle & Email Marketing (`lifecycle-email-marketing`)**: `subagent-lifecycle-email-specialist`
161
+
162
+ #### 🔒 Security Operations
163
+ - **Lead Orchestrator**: `orchestrator-security`
164
+ - **Essentials Base (`security-operations`)**: `subagent-security-engineer`
165
+
166
+ #### 🔎 Deep Technical Research
167
+ - **Lead Orchestrator**: `orchestrator-research`
168
+ - **Essentials Base (`deep-research`)**: `subagent-deep-research`, `subagent-socratic-mentor`, `subagent-repo-index`
169
+
170
+ #### 💞 Business Strategy & Economics
171
+ - **Lead Orchestrator**: `orchestrator-business`
172
+ - **Essentials Base (`business-strategy`)**: `subagent-business-panel-experts`
173
+
174
+ #### ðŸĒ Organization Bundles (Tier 2 / Experimental)
175
+ - **Digital Product Agency (`digital-agency` 🚧)**: *(Currently under development)*
176
+
177
+ </details>
113
178
 
114
179
  ---
115
180
 
@@ -145,9 +210,6 @@ agents add software-engineering
145
210
  agents add mobile-development
146
211
  agents add ai-ml-engineering
147
212
 
148
- # Install organization bundle in Brainstorming mode (ideation only, skips live MCP checks)
149
- agents add digital-agency --mode brainstorming
150
-
151
213
  # Install whole department domain
152
214
  agents add domain:engineering
153
215
  agents add domain:marketing
@@ -450,46 +512,113 @@ When contributing new skills or adapting external skills:
450
512
 
451
513
  Agents United proudly builds upon, adapts, and integrates contributions from creators across the open AI agent, cloud platform, and developer tooling ecosystem:
452
514
 
453
- - **Matt Pocock** ([@mattpocock](https://github.com/mattpocock) / [mattpocock/skills](https://github.com/mattpocock/skills)):
454
- - **`/grill-with-docs`** & **`/grill-me`**: Socratic alignment grilling, requirements clarification, and ADR authoring.
455
- - **`/domain-modeling`**: Ubiquitous language definition and `CONTEXT.md` domain dictionary maintenance.
456
- - **`/to-spec`** & **`/to-tickets`**: PRD/spec generation and task ticket decomposition.
457
- - **`/diagnosing-bugs`**: Evidence-driven bug diagnosis and root-cause analysis.
458
- - **`/git-guardrails`**: Version control safety rules and protection policies.
459
- - **`/handoff`**: Session progress persistence and context handoff notes.
460
- - **Modal Labs** ([@modal-labs](https://github.com/modal-labs) / [modal.com](https://modal.com)):
461
- - **`modal-serverless-python`**: Serverless Python execution, container image definition, and GPU acceleration patterns.
462
- - **Replicate** ([@replicate](https://github.com/replicate) / [replicate.com](https://replicate.com)):
463
- - **`replicate-model-inference`**: Hosted machine learning model inference API, webhook callbacks, and prediction streaming.
464
- - **RunPod** ([@runpod](https://github.com/runpod) / [runpod.io](https://runpod.io)):
465
- - **`runpod-gpu-orchestration`**: Serverless GPU handler architecture and containerized worker management.
466
- - **Ollama & vLLM Teams** ([ollama.com](https://ollama.com) / [vllm.ai](https://vllm.ai)):
467
- - **`local-llm-inference`**: Self-hosted LLM execution, quantization formats, and high-throughput PagedAttention serving.
468
- - **LangChain & LlamaIndex** ([langchain.com](https://langchain.com) / [llamaindex.ai](https://llamaindex.ai)):
469
- - **`rag-vector-pipeline`**: Retrieval-Augmented Generation architectures, hybrid retrieval, and re-ranking pipelines.
470
- - **Hugging Face** ([@huggingface](https://github.com/huggingface) / [huggingface.co](https://huggingface.co)):
471
- - **`hf-model-evaluation`**: Model benchmarking, Evaluate metrics, and model scorecard methodologies.
472
- - **Qdrant, Pinecone & Chroma** ([qdrant.tech](https://qdrant.tech) / [pinecone.io](https://pinecone.io) / [trychroma.com](https://www.trychroma.com)):
473
- - **`vector-database-design`**: Production vector indexing, HNSW graph tuning, and payload filtering.
474
- - **Vercel Engineering** ([@vercel](https://github.com/vercel) / [vercel.com](https://vercel.com)):
475
- - **`vercel-deploy-best-practices`**: Edge Middleware routing, Server Actions, and Preview environments.
476
- - **`react-best-practices`**: Next.js App Router, Server Components & Core Web Vitals optimization.
477
- - **Lovable, v0 (Vercel) & Bolt (StackBlitz)** ([lovable.dev](https://lovable.dev) / [v0.dev](https://v0.dev) / [bolt.new](https://bolt.new)):
478
- - **`ai-prototype-refactoring`**: Ingestion and modularization methodologies for rapid AI-generated frontend prototypes.
479
- - **Supabase** ([@supabase](https://github.com/supabase) / [supabase.com](https://supabase.com)):
480
- - **`supabase-backend-architecture`**: PostgreSQL database design, Row Level Security (RLS), Edge Functions, and Realtime sync.
481
- - **ChiselStrike & Turso Community** ([@tursodatabase](https://github.com/tursodatabase) / [turso.tech](https://turso.tech)):
482
- - **`turso-distributed-sqlite`**: LibSQL distributed SQLite, embedded replicas with auto-sync, and database branching.
483
- - **Microsoft Azure Community** ([learn.microsoft.com/azure/bicep](https://learn.microsoft.com/azure/azure-resource-manager/bicep)):
484
- - **`azure-infrastructure-bicep`**: Enterprise Bicep Infrastructure-as-Code, Azure Container Apps (ACA), and Azure OpenAI private networking.
485
- - **Currents & Microsoft Playwright Community** ([currents-dev/playwright-best-practices-skill](https://github.com/currents-dev/playwright-best-practices-skill)):
486
- - **`playwright-best-practices`**: Resilient Page Object Models and deterministic auto-waiting browser tests.
487
- - **wshobson** ([wshobson/agents](https://github.com/wshobson/agents)):
488
- - **`mobile-ios-design`** & **`mobile-android-design`**: SwiftUI & Jetpack Compose design system patterns.
489
- - **Salesforce** ([forcedotcom/sf-skills](https://github.com/forcedotcom/sf-skills)):
490
- - **`mobile-platform-offline-validate`**: Offline-first local database caching and conflict resolution.
491
- - **tovimx** ([tovimx/maestro-mobile-testing-skill](https://github.com/tovimx/maestro-mobile-testing-skill)):
492
- - **`maestro-mobile-testing`**: Declarative cross-platform mobile UI test automation.
515
+ <details>
516
+ <summary><strong>Matt Pocock (<a href="https://github.com/mattpocock">@mattpocock</a> / <a href="https://github.com/mattpocock/skills">mattpocock/skills</a>)</strong></summary>
517
+
518
+ - **`/grill-with-docs`** & **`/grill-me`**: Socratic alignment grilling, requirements clarification, and ADR authoring.
519
+ - **`/domain-modeling`**: Ubiquitous language definition and `CONTEXT.md` domain dictionary maintenance.
520
+ - **`/to-spec`** & **`/to-tickets`**: PRD/spec generation and task ticket decomposition.
521
+ - **`/diagnosing-bugs`**: Evidence-driven bug diagnosis and root-cause analysis.
522
+ - **`/git-guardrails`**: Version control safety rules and protection policies.
523
+ - **`/handoff`**: Session progress persistence and context handoff notes.
524
+ </details>
525
+
526
+ <details>
527
+ <summary><strong>Modal Labs (<a href="https://github.com/modal-labs">@modal-labs</a> / <a href="https://modal.com">modal.com</a>)</strong></summary>
528
+
529
+ - **`modal-serverless-python`**: Serverless Python execution, container image definition, and GPU acceleration patterns.
530
+ </details>
531
+
532
+ <details>
533
+ <summary><strong>Replicate (<a href="https://github.com/replicate">@replicate</a> / <a href="https://replicate.com">replicate.com</a>)</strong></summary>
534
+
535
+ - **`replicate-model-inference`**: Hosted machine learning model inference API, webhook callbacks, and prediction streaming.
536
+ </details>
537
+
538
+ <details>
539
+ <summary><strong>RunPod (<a href="https://github.com/runpod">@runpod</a> / <a href="https://runpod.io">runpod.io</a>)</strong></summary>
540
+
541
+ - **`runpod-gpu-orchestration`**: Serverless GPU handler architecture and containerized worker management.
542
+ </details>
543
+
544
+ <details>
545
+ <summary><strong>Ollama & vLLM Teams (<a href="https://ollama.com">ollama.com</a> / <a href="https://vllm.ai">vllm.ai</a>)</strong></summary>
546
+
547
+ - **`local-llm-inference`**: Self-hosted LLM execution, quantization formats, and high-throughput PagedAttention serving.
548
+ </details>
549
+
550
+ <details>
551
+ <summary><strong>LangChain & LlamaIndex (<a href="https://langchain.com">langchain.com</a> / <a href="https://llamaindex.ai">llamaindex.ai</a>)</strong></summary>
552
+
553
+ - **`rag-vector-pipeline`**: Retrieval-Augmented Generation architectures, hybrid retrieval, and re-ranking pipelines.
554
+ </details>
555
+
556
+ <details>
557
+ <summary><strong>Hugging Face (<a href="https://github.com/huggingface">@huggingface</a> / <a href="https://huggingface.co">huggingface.co</a>)</strong></summary>
558
+
559
+ - **`hf-model-evaluation`**: Model benchmarking, Evaluate metrics, and model scorecard methodologies.
560
+ </details>
561
+
562
+ <details>
563
+ <summary><strong>Qdrant, Pinecone & Chroma (<a href="https://qdrant.tech">qdrant.tech</a> / <a href="https://pinecone.io">pinecone.io</a> / <a href="https://www.trychroma.com">trychroma.com</a>)</strong></summary>
564
+
565
+ - **`vector-database-design`**: Production vector indexing, HNSW graph tuning, and payload filtering.
566
+ </details>
567
+
568
+ <details>
569
+ <summary><strong>Vercel Engineering (<a href="https://github.com/vercel">@vercel</a> / <a href="https://vercel.com">vercel.com</a>)</strong></summary>
570
+
571
+ - **`vercel-deploy-best-practices`**: Edge Middleware routing, Server Actions, and Preview environments.
572
+ - **`react-best-practices`**: Next.js App Router, Server Components & Core Web Vitals optimization.
573
+ </details>
574
+
575
+ <details>
576
+ <summary><strong>Lovable, v0 (Vercel) & Bolt (StackBlitz) (<a href="https://lovable.dev">lovable.dev</a> / <a href="https://v0.dev">v0.dev</a> / <a href="https://bolt.new">bolt.new</a>)</strong></summary>
577
+
578
+ - **`ai-prototype-refactoring`**: Ingestion and modularization methodologies for rapid AI-generated frontend prototypes.
579
+ </details>
580
+
581
+ <details>
582
+ <summary><strong>Supabase (<a href="https://github.com/supabase">@supabase</a> / <a href="https://supabase.com">supabase.com</a>)</strong></summary>
583
+
584
+ - **`supabase-backend-architecture`**: PostgreSQL database design, Row Level Security (RLS), Edge Functions, and Realtime sync.
585
+ </details>
586
+
587
+ <details>
588
+ <summary><strong>ChiselStrike & Turso Community (<a href="https://github.com/tursodatabase">@tursodatabase</a> / <a href="https://turso.tech">turso.tech</a>)</strong></summary>
589
+
590
+ - **`turso-distributed-sqlite`**: LibSQL distributed SQLite, embedded replicas with auto-sync, and database branching.
591
+ </details>
592
+
593
+ <details>
594
+ <summary><strong>Microsoft Azure Community (<a href="https://learn.microsoft.com/azure/bicep">learn.microsoft.com/azure/bicep</a>)</strong></summary>
595
+
596
+ - **`azure-infrastructure-bicep`**: Enterprise Bicep Infrastructure-as-Code, Azure Container Apps (ACA), and Azure OpenAI private networking.
597
+ </details>
598
+
599
+ <details>
600
+ <summary><strong>Currents & Microsoft Playwright Community (<a href="https://github.com/currents-dev/playwright-best-practices-skill">currents-dev/playwright-best-practices-skill</a>)</strong></summary>
601
+
602
+ - **`playwright-best-practices`**: Resilient Page Object Models and deterministic auto-waiting browser tests.
603
+ </details>
604
+
605
+ <details>
606
+ <summary><strong>wshobson (<a href="https://github.com/wshobson/agents">wshobson/agents</a>)</strong></summary>
607
+
608
+ - **`mobile-ios-design`** & **`mobile-android-design`**: SwiftUI & Jetpack Compose design system patterns.
609
+ </details>
610
+
611
+ <details>
612
+ <summary><strong>Salesforce (<a href="https://github.com/forcedotcom/sf-skills">forcedotcom/sf-skills</a>)</strong></summary>
613
+
614
+ - **`mobile-platform-offline-validate`**: Offline-first local database caching and conflict resolution.
615
+ </details>
616
+
617
+ <details>
618
+ <summary><strong>tovimx (<a href="https://github.com/tovimx/maestro-mobile-testing-skill">tovimx/maestro-mobile-testing-skill</a>)</strong></summary>
619
+
620
+ - **`maestro-mobile-testing`**: Declarative cross-platform mobile UI test automation.
621
+ </details>
493
622
 
494
623
  ---
495
624
 
package/dist/cli.js CHANGED
@@ -209,7 +209,7 @@ import path2 from "path";
209
209
  var HOST_REGISTRY = {
210
210
  agents: {
211
211
  id: "agents",
212
- label: "Antigravity \u2014 main library (.agents/)",
212
+ label: "Google Antigravity & Master Library (\u{1F4C1} .agents/)",
213
213
  projectDir: ".agents",
214
214
  globalDirSegments: [".agents"],
215
215
  agentsSubdir: "agents",
@@ -219,7 +219,7 @@ var HOST_REGISTRY = {
219
219
  },
220
220
  gemini: {
221
221
  id: "gemini",
222
- label: "Antigravity legacy (.gemini/)",
222
+ label: "Antigravity legacy (\u{1F4C1} .gemini/)",
223
223
  projectDir: ".gemini",
224
224
  globalDirSegments: [".gemini", "config"],
225
225
  agentsSubdir: "agents",
@@ -229,7 +229,7 @@ var HOST_REGISTRY = {
229
229
  },
230
230
  claude: {
231
231
  id: "claude",
232
- label: "Claude Code (.claude/)",
232
+ label: "Anthropic Claude Code (\u{1F4C1} .claude/)",
233
233
  projectDir: ".claude",
234
234
  globalDirSegments: [".claude"],
235
235
  agentsSubdir: "agents",
@@ -239,7 +239,7 @@ var HOST_REGISTRY = {
239
239
  },
240
240
  cursor: {
241
241
  id: "cursor",
242
- label: "Cursor (.cursor/)",
242
+ label: "Cursor IDE (\u{1F4C1} .cursor/)",
243
243
  projectDir: ".cursor",
244
244
  globalDirSegments: [".cursor"],
245
245
  agentsSubdir: "agents",
@@ -249,7 +249,7 @@ var HOST_REGISTRY = {
249
249
  },
250
250
  cline: {
251
251
  id: "cline",
252
- label: "Cline (.cline/)",
252
+ label: "Cline (\u{1F4C1} .cline/)",
253
253
  projectDir: ".cline",
254
254
  globalDirSegments: [".cline"],
255
255
  agentsSubdir: "agents",
@@ -259,7 +259,7 @@ var HOST_REGISTRY = {
259
259
  },
260
260
  opencode: {
261
261
  id: "opencode",
262
- label: "OpenCode (.opencode/)",
262
+ label: "OpenCode (\u{1F4C1} .opencode/)",
263
263
  projectDir: ".opencode",
264
264
  globalDirSegments: [".config", "opencode"],
265
265
  agentsSubdir: "agent",
@@ -269,7 +269,7 @@ var HOST_REGISTRY = {
269
269
  },
270
270
  codex: {
271
271
  id: "codex",
272
- label: "Codex & AGENTS.md tools (AGENTS.md)",
272
+ label: "OpenAI Codex, Copilot, Aider & Zed (\u{1F4C4} ./AGENTS.md)",
273
273
  projectDir: ".",
274
274
  globalDirSegments: [".codex"],
275
275
  agentsSubdir: void 0,
@@ -2630,42 +2630,42 @@ cli.command("add [identifier]", "Add a bundle, agent, skill, or workflow to proj
2630
2630
  );
2631
2631
  }
2632
2632
  const hostSelection = await multiselect({
2633
- message: "1. Which AI assistants will you work with?",
2633
+ message: "1. Which AI tools should we configure agent teams & personas for?",
2634
2634
  options: [
2635
2635
  {
2636
2636
  value: "agents",
2637
2637
  label: HOST_REGISTRY.agents.label,
2638
- hint: detectedHosts.includes("agents") ? "found in this project" : "the one folder you edit \u2014 always included"
2639
- },
2640
- {
2641
- value: "gemini",
2642
- label: HOST_REGISTRY.gemini.label,
2643
- hint: detectedHosts.includes("gemini") ? "found in this project" : "older Antigravity folder"
2638
+ hint: detectedHosts.includes("agents") ? "found in this project" : "the master library where all skills & agents live \u2014 always included"
2644
2639
  },
2645
2640
  {
2646
2641
  value: "claude",
2647
2642
  label: HOST_REGISTRY.claude.label,
2648
- hint: detectedHosts.includes("claude") ? "found in this project" : "gets its own translated copies"
2643
+ hint: detectedHosts.includes("claude") ? "found in this project" : "orchestrator & subagent personas for Claude Code"
2649
2644
  },
2650
2645
  {
2651
2646
  value: "cursor",
2652
2647
  label: HOST_REGISTRY.cursor.label,
2653
- hint: detectedHosts.includes("cursor") ? "found in this project" : "gets its own translated copies"
2648
+ hint: detectedHosts.includes("cursor") ? "found in this project" : "orchestrator & subagent personas for Cursor IDE"
2654
2649
  },
2655
2650
  {
2656
2651
  value: "cline",
2657
2652
  label: HOST_REGISTRY.cline.label,
2658
- hint: detectedHosts.includes("cline") ? "found in this project" : "gets its own translated copies"
2653
+ hint: detectedHosts.includes("cline") ? "found in this project" : "roles, skills, coordinator rules & team manifests for Cline"
2659
2654
  },
2660
2655
  {
2661
2656
  value: "opencode",
2662
2657
  label: HOST_REGISTRY.opencode.label,
2663
- hint: detectedHosts.includes("opencode") ? "found in this project" : "gets its own translated copies"
2658
+ hint: detectedHosts.includes("opencode") ? "found in this project" : "orchestrator & subagent personas for OpenCode"
2664
2659
  },
2665
2660
  {
2666
2661
  value: "codex",
2667
2662
  label: HOST_REGISTRY.codex.label,
2668
- hint: detectedHosts.includes("codex") ? "found in this project" : "an index file many tools read"
2663
+ hint: detectedHosts.includes("codex") ? "found in this project" : "root index linking orchestrators, subagents & skills for Codex, Copilot, Aider & Zed"
2664
+ },
2665
+ {
2666
+ value: "gemini",
2667
+ label: HOST_REGISTRY.gemini.label,
2668
+ hint: detectedHosts.includes("gemini") ? "found in this project" : "older Antigravity 1.0 / Gemini folder"
2669
2669
  }
2670
2670
  ],
2671
2671
  initialValues: detectedHosts.length > 0 ? detectedHosts : ["agents"],
@@ -2677,8 +2677,8 @@ cli.command("add [identifier]", "Add a bundle, agent, skill, or workflow to proj
2677
2677
  fanout = Array.from(/* @__PURE__ */ new Set([...fanout, ...wizardPlan.fanout]));
2678
2678
  if (wizardPlan.addedCanonicalStore) {
2679
2679
  note(
2680
- "Added .agents/ \u2014 the main library your assistants share.\nEach assistant you picked gets its own translated copy. Edit only .agents/.",
2681
- "Main library"
2680
+ "Added .agents/ \u2014 the master repository library your tools share.\nEach tool you selected gets its own orchestrators and subagents. Edit only .agents/.",
2681
+ "Master library"
2682
2682
  );
2683
2683
  }
2684
2684
  }