agentbrief 0.1.0 → 0.1.2

package/README.md

@@ -11,20 +11,47 @@
 One command turns your Claude Code / Cursor / OpenCode / Codex from a generic assistant into a specialized professional.
 
 ```bash
-npx agentbrief use security-auditor
-# Your AI coding agent is now an OWASP security auditor
+npx agentbrief use fullstack-dev
+# Your agent now enforces strict TypeScript, follows Next.js conventions,
+# builds accessible UIs, and reviews PRs like a staff engineer — 8 skills in 1 command
 ```
 
 ## What It Does
 
-AgentBrief compiles a **brief** (role definition + domain knowledge + behavioral rules) into the instruction files your AI agent reads — `CLAUDE.md`, `.cursorrules`, `AGENTS.md` — with content automatically adapted for each engine.
+AgentBrief compiles a **brief** (role definition + domain knowledge + executable skills) into the instruction files your AI agent reads — `CLAUDE.md`, `.cursorrules`, `AGENTS.md` — with content automatically adapted for each engine.
 
-```
-Before: Generic AI assistant that knows nothing about your domain
-After:   Security auditor that cites CWE numbers, checks OWASP Top 10,
-         and refuses to approve code with injection vulnerabilities
+<details>
+<summary><b>See it in action</b> — what happens when you run <code>agentbrief use fullstack-dev</code></summary>
+
+Your `CLAUDE.md` gets this injected (existing content preserved):
+
+```markdown
+<!-- agentbrief:fullstack-dev:start -->
+# AgentBrief: fullstack-dev
+
+## Role
+You are a senior full-stack TypeScript developer. You build production
+applications with Next.js 15, React 19, and Tailwind CSS. You enforce
+strict type safety and review your own code with principal-engineer rigor.
+
+## Constraints
+- Never use `any` — always annotate return types on exports
+- Server Components by default — only add 'use client' when needed
+- WCAG 2.1 AA minimum — semantic HTML, keyboard navigation
+
+## Skills
+- **next-best-practices** — USE WHEN: Writing Next.js code
+- **typescript-advanced-types** — USE WHEN: Complex type logic
+- **architecture-review** — USE WHEN: Reviewing PRs
+- **design-review-checklist** — USE WHEN: Checking UI quality
+- **agent-browser** — USE WHEN: Visual verification needed
+  ...and 4 more
+<!-- agentbrief:fullstack-dev:end -->
 ```
 
+Cleanly removable with `agentbrief eject fullstack-dev`.
+</details>
+
 - **One command to apply, one to remove** — `use` and `eject`
 - **Non-invasive** — injected content is wrapped in markers, your existing files are untouched
 - **Stackable** — apply multiple briefs; later ones layer on top
@@ -70,33 +97,36 @@ agentbrief init my-agent --template security
 
 ## Official Registry
 
-```bash
-agentbrief search
-```
+**⚡ Quick Start Packs** — best if you're not sure where to begin:
 
-```
-Available briefs:
-
-  NAME               TRUST      DESCRIPTION
-  security-auditor   official   OWASP/CWE security review specialist
-  code-reviewer      official   Rigorous PR review — naming, tests, architecture
-  typescript-strict  official   TypeScript type safety guardian — zero any
-  nextjs-fullstack   official   Next.js 15 + App Router + React 19 + Tailwind
-  frontend-design    official   React + Tailwind + shadcn/ui design engineering
-  devops-sre         official   Infrastructure monitoring, incident response, IaC
-  tech-writer        official   Technical documentation with style guide adherence
-  growth-engineer    official   CRO, SEO, analytics, growth engineering
-  product-manager    official   PRD generation, user stories, prioritization
-  startup-builder    official   Idea validation → MVP → launch workflow
-  qa-engineer        official   Automated QA — find bugs, write tests, fix with atomic commits
-  data-analyst       official   Business intelligence — metrics, SQL, dashboards, data storytelling
-  fullstack-dev      official   Full-stack TypeScript developer — extends 4 briefs, 8 skills
-  startup-kit        official   Startup builder kit — extends 4 briefs, 9 skills
-```
+| Pack | What your agent gains |
+|------|----------------------|
+| `fullstack-dev` | Strict TypeScript + Next.js conventions + accessible UI + PR reviews (9 skills) |
+| `startup-kit` | Product specs + SEO audit + growth analytics + security review + launch strategy (12 skills) |
+
+**Code Quality & Engineering:**
+
+| Brief | Your agent becomes... |
+|-------|----------------------|
+| `security-auditor` | OWASP security reviewer who cites CWE numbers |
+| `code-reviewer` | Staff engineer who catches architecture + logic issues |
+| `qa-engineer` | QA who finds bugs, writes tests, fixes with atomic commits |
+| `typescript-strict` | Type safety enforcer — zero `any`, exhaustive checks |
+| `nextjs-fullstack` | Next.js 15 specialist (App Router, RSC, Tailwind) |
+| `frontend-design` | Design engineer with 80-item review checklist |
+| `devops-sre` | SRE who sets up CI/CD, monitoring, incident response |
+| `tech-writer` | Documentation specialist with API docs + release notes |
 
-Just type `agentbrief use <name>` — no need for full GitHub URLs.
+**Product, Growth & Business:**
 
-Browse the **[full Catalog](./CATALOG.md)** for more roles across development, product, marketing, operations, finance, legal, and startup — with links to community resources.
+| Brief | Your agent becomes... |
+|-------|----------------------|
+| `product-manager` | PM who writes PRDs with RICE/ICE prioritization |
+| `growth-engineer` | Growth hacker with SEO audit + analytics + content strategy |
+| `data-analyst` | BI analyst with metrics frameworks + SQL patterns |
+| `startup-builder` | Startup advisor with CEO review + launch planning |
+
+Browse the **[full Catalog](./CATALOG.md)** or the **[Website](https://0xranx.github.io/agentbrief)** to see details for each brief.
 
 ## What's a Brief?
 
@@ -114,7 +144,19 @@ my-brief/
         └── ...         #   any supporting files
 ```
 
-See `examples/security-auditor/` for a complete example. Read the **[Authoring Guide](./docs/AUTHORING.md)** to create your own.
+See `examples/security-auditor/` for a complete example.
+
+## Create & Publish Your Own
+
+Share your expertise as a reusable brief:
+
+1. **Scaffold** — `agentbrief init my-agent` (or `--template security`)
+2. **Customize** — Edit `personality.md`, add `knowledge/` and `skills/`
+3. **Test** — `agentbrief use ./my-agent` in a temp directory
+4. **Preview** — `agentbrief preview ./my-agent` to see compiled output
+5. **Publish** — Push to GitHub → `agentbrief use github:you/my-agent`
+
+Want it in the official registry? [Submit a PR](./CONTRIBUTING.md). Read the full **[Authoring Guide](https://0xranx.github.io/agentbrief/docs.html#AUTHORING)**.
 
 ## Supported Engines
 
@@ -136,6 +178,11 @@ AgentBrief compiles each brief with optimizations for the target engine:
 
 Commit the engine instruction files (`CLAUDE.md`, `.cursorrules`, `AGENTS.md`) so your team shares the same agent behavior.
 
+## Community
+
+- [GitHub Discussions](https://github.com/0xranx/agentbrief/discussions) — Questions, showcase your briefs, feature ideas
+- [Contributing Guide](./CONTRIBUTING.md) — How to submit briefs to the registry
+
 ## License
 
 [MIT](./LICENSE)

package/briefs/base-agent/brief.yaml

@@ -0,0 +1,6 @@
+name: base-agent
+version: "1.0.0"
+description: Foundation layer — cross-cutting skills every AI coding agent should have
+personality: personality.md
+skills:
+  - skills/

package/briefs/base-agent/personality.md

@@ -0,0 +1,9 @@
+## Role
+
+You are an AI coding agent with foundational capabilities for browsing, inspecting, and interacting with web applications and services.
+
+## Constraints
+
+- Always use agent-browser for visual verification before claiming UI work is complete
+- Take screenshots to document visual state when reporting issues
+- Use snapshots (accessibility tree) instead of raw DOM for element interaction