azrole 3.0.0
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.
- package/README.md +548 -0
- package/bin/cli.js +561 -0
- package/docs/case-studies/book-manuscript.md +301 -0
- package/package.json +48 -0
- package/templates/agents/orchestrator.md +1868 -0
- package/templates/commands/dream.md +13 -0
- package/templates/commands/evolve.md +3 -0
- package/templates/commands/explain.md +13 -0
- package/templates/commands/fix.md +15 -0
- package/templates/commands/level-up.md +3 -0
- package/templates/commands/setup.md +26 -0
- package/templates/commands/ship.md +14 -0
- package/templates/commands/status.md +14 -0
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
I want to build a project. Here's my idea:
|
|
2
|
+
|
|
3
|
+
$ARGUMENTS
|
|
4
|
+
|
|
5
|
+
Orchestrator — analyze this project idea, detect the tech stack, scan my current
|
|
6
|
+
environment level (0-10), then build everything I need progressively:
|
|
7
|
+
CLAUDE.md, MCP config, skills, commands, memory system, and specialized dev agents.
|
|
8
|
+
|
|
9
|
+
Show me progress at each level and what was created when done.
|
|
10
|
+
|
|
11
|
+
If the project idea above is blank or empty, ask me: "What do you want to build? Give me your idea and the technologies you want to use."
|
|
12
|
+
|
|
13
|
+
If no tech stack was mentioned above, ask me ONE question: "What technologies do you want to use?"
|
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
Explain this to me like I'm not a developer:
|
|
2
|
+
|
|
3
|
+
$ARGUMENTS
|
|
4
|
+
|
|
5
|
+
Rules:
|
|
6
|
+
- Use simple, everyday language — no jargon
|
|
7
|
+
- If I pasted code, explain what it does step by step
|
|
8
|
+
- If I pasted an error, explain what went wrong and how to fix it
|
|
9
|
+
- If I named a file or folder, read it and explain what it's for
|
|
10
|
+
- If I asked a concept question, explain it with an analogy
|
|
11
|
+
- Keep it short — 2-3 paragraphs max unless I ask for more detail
|
|
12
|
+
|
|
13
|
+
If nothing was provided above, ask: "What would you like me to explain? You can paste code, an error message, a file name, or ask about any concept."
|
|
@@ -0,0 +1,15 @@
|
|
|
1
|
+
Something is broken. Here's what I see:
|
|
2
|
+
|
|
3
|
+
$ARGUMENTS
|
|
4
|
+
|
|
5
|
+
Find the bug, fix it, and explain what went wrong in plain English.
|
|
6
|
+
|
|
7
|
+
Steps:
|
|
8
|
+
1. Read the error message or description above
|
|
9
|
+
2. Search the codebase for the relevant files
|
|
10
|
+
3. Identify the root cause
|
|
11
|
+
4. Fix it
|
|
12
|
+
5. Verify the fix works (run the app/tests if possible)
|
|
13
|
+
6. Explain what happened and what you changed — assume I'm not a developer
|
|
14
|
+
|
|
15
|
+
If no error or description was provided above, ask: "What's going wrong? Paste the error or describe what you expected to happen."
|
|
@@ -0,0 +1,26 @@
|
|
|
1
|
+
Scan my existing project and build the AI coding environment for it.
|
|
2
|
+
|
|
3
|
+
Orchestrator — this is an EXISTING project with code already written. Do NOT ask what I want to build. Instead:
|
|
4
|
+
|
|
5
|
+
1. Scan the current directory thoroughly:
|
|
6
|
+
- Read package.json, pyproject.toml, requirements.txt, Cargo.toml, go.mod, etc.
|
|
7
|
+
- Look at the directory structure to understand the architecture
|
|
8
|
+
- Read any existing project rules file (CLAUDE.md, AGENTS.md, GEMINI.md, .cursorrules)
|
|
9
|
+
- Check for existing config dirs (.claude/, .gemini/, .opencode/, .codex/, .cursor/)
|
|
10
|
+
- Identify the tech stack, frameworks, and patterns from actual code
|
|
11
|
+
|
|
12
|
+
2. Run the environment scanner to detect my current level (0-10)
|
|
13
|
+
|
|
14
|
+
3. Build everything I'm missing up to Level 5:
|
|
15
|
+
- Project rules file based on what actually exists in the codebase
|
|
16
|
+
- MCP servers for the detected technologies
|
|
17
|
+
- Skills based on the actual patterns found in the code
|
|
18
|
+
- Memory system with a real codebase map
|
|
19
|
+
- Specialized agents matched to the actual directory structure
|
|
20
|
+
|
|
21
|
+
4. Show me what was detected and what was built
|
|
22
|
+
|
|
23
|
+
If $ARGUMENTS is provided, use it as additional context about the project.
|
|
24
|
+
Otherwise, figure everything out from the code.
|
|
25
|
+
|
|
26
|
+
$ARGUMENTS
|
|
@@ -0,0 +1,14 @@
|
|
|
1
|
+
Save my work and push it to GitHub.
|
|
2
|
+
|
|
3
|
+
Steps:
|
|
4
|
+
1. Run `git status` to see what changed
|
|
5
|
+
2. If this isn't a git repo yet, run `git init` and create an initial commit
|
|
6
|
+
3. Stage all changed files (but NEVER stage .env, credentials, secrets, or API keys)
|
|
7
|
+
4. Generate a clear commit message describing what changed
|
|
8
|
+
5. Commit the changes
|
|
9
|
+
6. If a remote is set up, push to the current branch
|
|
10
|
+
7. If no remote exists, tell me how to connect to GitHub
|
|
11
|
+
|
|
12
|
+
Show me a summary of what was shipped.
|
|
13
|
+
|
|
14
|
+
If there are no changes to ship, say: "Nothing to ship — your code is already up to date."
|
|
@@ -0,0 +1,14 @@
|
|
|
1
|
+
Give me a quick overview of my project right now.
|
|
2
|
+
|
|
3
|
+
Check and report:
|
|
4
|
+
1. **Project health** — does the app run? Any obvious errors?
|
|
5
|
+
- Try to detect the start command (package.json scripts, pyproject.toml, Makefile, etc.)
|
|
6
|
+
- Check for common issues (missing dependencies, env vars, config files)
|
|
7
|
+
|
|
8
|
+
2. **What's changed** — run `git status` and `git diff --stat` to show recent changes
|
|
9
|
+
|
|
10
|
+
3. **Current level** — if .devteam/blueprint.json exists, scan the Claude Code environment level (0-10)
|
|
11
|
+
|
|
12
|
+
4. **Next steps** — based on what you see, suggest 2-3 things I could work on next
|
|
13
|
+
|
|
14
|
+
Keep the output short and visual. Use checkmarks and X marks. No walls of text.
|