@0xtiby/toby 0.0.1
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 +72 -0
- package/dist/cli.js +3177 -0
- package/package.json +42 -0
- package/prompts/PROMPT_BUILD.md +94 -0
- package/prompts/PROMPT_PLAN.md +124 -0
- package/prompts/prompts.test.ts +76 -0
package/README.md
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
1
|
+
# toby
|
|
2
|
+
|
|
3
|
+
> Turn markdown specs into working code with AI-powered plan and build loops
|
|
4
|
+
|
|
5
|
+
[](https://www.npmjs.com/package/@0xtiby/toby)
|
|
6
|
+
[](LICENSE)
|
|
7
|
+
|
|
8
|
+
## What is toby?
|
|
9
|
+
|
|
10
|
+
Toby is a CLI tool that turns markdown specifications into working code through AI-driven plan and build loops. You write specs describing what you want, and toby orchestrates an AI agent to plan the implementation, then iteratively builds it — discovering files, generating code, and running validation until the spec is complete.
|
|
11
|
+
|
|
12
|
+
## How it works
|
|
13
|
+
|
|
14
|
+
Toby follows a three-phase loop:
|
|
15
|
+
|
|
16
|
+
1. **Spec** — You write a markdown file describing what you want built (features, acceptance criteria, constraints)
|
|
17
|
+
2. **Plan** — Toby sends your spec to an AI CLI, which analyzes the codebase and produces an implementation plan with concrete tasks
|
|
18
|
+
3. **Build** — Toby iteratively executes each task through the AI CLI, running your validation command between iterations until the spec is complete
|
|
19
|
+
|
|
20
|
+
Each phase feeds into the next: specs drive plans, plans drive builds, and build results update the project status so you always know where things stand.
|
|
21
|
+
|
|
22
|
+
## Quick start
|
|
23
|
+
|
|
24
|
+
```bash
|
|
25
|
+
# Install toby globally
|
|
26
|
+
npm install -g @0xtiby/toby
|
|
27
|
+
|
|
28
|
+
# Initialize toby in your project
|
|
29
|
+
toby init
|
|
30
|
+
|
|
31
|
+
# Write a spec in your specs directory (default: specs/)
|
|
32
|
+
# e.g. specs/add-auth.md describing the feature you want
|
|
33
|
+
|
|
34
|
+
# Generate an implementation plan from your spec
|
|
35
|
+
toby plan --spec=add-auth
|
|
36
|
+
|
|
37
|
+
# Build the planned spec with AI
|
|
38
|
+
toby build --spec=add-auth
|
|
39
|
+
```
|
|
40
|
+
|
|
41
|
+
- `toby init` — sets up configuration and specs directory in your project
|
|
42
|
+
- `toby plan` — sends a spec to the AI CLI to produce an implementation plan
|
|
43
|
+
- `toby build` — iteratively executes the plan, running validation between iterations
|
|
44
|
+
- `toby status` — shows progress across all specs
|
|
45
|
+
- `toby config` — manage CLI, model, and project settings
|
|
46
|
+
|
|
47
|
+
## Development
|
|
48
|
+
|
|
49
|
+
```bash
|
|
50
|
+
# Install dependencies
|
|
51
|
+
pnpm install
|
|
52
|
+
|
|
53
|
+
# Build the project
|
|
54
|
+
pnpm build
|
|
55
|
+
|
|
56
|
+
# Link globally for local testing
|
|
57
|
+
pnpm link --global
|
|
58
|
+
|
|
59
|
+
# Unlink when done
|
|
60
|
+
pnpm unlink --global @0xtiby/toby
|
|
61
|
+
```
|
|
62
|
+
|
|
63
|
+
## Documentation
|
|
64
|
+
|
|
65
|
+
- [CLI & Config Reference](docs/reference.md) — all commands, flags, and config.json options
|
|
66
|
+
- [Prompt Authoring Guide](docs/prompts.md) — how to write and customize prompt templates
|
|
67
|
+
- [Writing Specs](docs/specs.md) — spec file format, naming conventions, and ordering
|
|
68
|
+
- [Sessions & Transcripts](docs/sessions.md) — session naming, crash recovery, and transcript recording
|
|
69
|
+
|
|
70
|
+
## License
|
|
71
|
+
|
|
72
|
+
MIT
|