create-expert 0.0.9 → 0.0.10

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 CHANGED
@@ -4,6 +4,8 @@ Interactive wizard to create Perstack Experts.
4
4
 
5
5
  ## Usage
6
6
 
7
+ ### Interactive Mode (TUI)
8
+
7
9
  ```bash
8
10
  # Create a new Expert project
9
11
  npx create-expert
@@ -12,6 +14,37 @@ npx create-expert
12
14
  npx create-expert my-expert "Add error handling for edge cases"
13
15
  ```
14
16
 
17
+ ### Headless Mode (CLI)
18
+
19
+ For automation and scripting, use the `--headless` flag:
20
+
21
+ ```bash
22
+ # Create a new Expert project (headless)
23
+ npx create-expert --headless \
24
+ --provider anthropic \
25
+ --model claude-sonnet-4-5 \
26
+ --description "A code reviewer that checks for TypeScript best practices"
27
+
28
+ # Improve an existing Expert (headless)
29
+ npx create-expert my-expert "Add web search capability" --headless
30
+ ```
31
+
32
+ ## CLI Options
33
+
34
+ | Option | Description | Default |
35
+ |--------|-------------|---------|
36
+ | `--headless` | Run without TUI wizard | `false` |
37
+ | `--json` | Output raw JSON events (only with `--headless`) | `false` |
38
+ | `--provider <provider>` | LLM provider (`anthropic`, `openai`, `google`) | `anthropic` |
39
+ | `--model <model>` | Model name | Provider default |
40
+ | `--runtime <runtime>` | Execution runtime (`default`, `cursor`, `claude-code`, `gemini`) | `default` |
41
+ | `--description <desc>` | Expert description (required in headless mode for new projects) | - |
42
+ | `--cwd <path>` | Working directory | Current directory |
43
+
44
+ ### Output Modes
45
+
46
+ In headless mode, the default output is human-readable with progress indicators and a final summary. Use `--json` for raw JSON event stream (useful for scripting and automation).
47
+
15
48
  ## What it does
16
49
 
17
50
  ### New Project Setup
@@ -20,25 +53,29 @@ npx create-expert my-expert "Add error handling for edge cases"
20
53
  - LLMs: Anthropic, OpenAI, Google (via environment variables)
21
54
  - Runtimes: Cursor, Claude Code, Gemini CLI
22
55
 
23
- 2. **Configures your environment**
56
+ 2. **Configures your environment** (interactive mode only)
24
57
  - Prompts for API keys if needed
25
58
  - Creates `.env` file with credentials
26
59
 
27
- 3. **Creates project files**
60
+ 3. **Creates project files** (only if they don't exist)
28
61
  - `AGENTS.md` - Instructions for AI agents working on this project
29
62
  - `perstack.toml` - Expert definitions and runtime config
63
+ - If files already exist, they are preserved and reused
30
64
 
31
65
  4. **Runs the Expert creation flow**
32
- - Asks what kind of Expert you want
33
- - Uses `perstack run` to create and test the Expert
34
- - Iterates until the Expert works correctly
66
+ - Interactive: Uses `perstack start` with TUI
67
+ - Headless: Uses `perstack run` for non-interactive execution
35
68
 
36
69
  ### Improvement Mode
37
70
 
38
71
  When called with an Expert name:
39
72
 
40
73
  ```bash
74
+ # Interactive
41
75
  npx create-expert my-expert "Add web search capability"
76
+
77
+ # Headless
78
+ npx create-expert my-expert "Add web search capability" --headless
42
79
  ```
43
80
 
44
81
  Skips project setup and goes straight to improvement, using the existing configuration.