Metadata-Version: 2.4
Name: tx-ai-agent-sdd
Version: 0.1.1
Summary: CLI scaffold generator for spec-driven AI agent documentation
Author: Troyhack
License-Expression: MIT
Keywords: spec-driven,ai-agent,sdd,scaffold,documentation
Classifier: Intended Audience :: Developers
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Operating System :: OS Independent
Classifier: Topic :: Software Development :: Documentation
Classifier: Topic :: Software Development :: Code Generators
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Dynamic: license-file

# tx-ai-agent-sdd

`tx-ai-agent-sdd` bootstraps a spec-driven workspace for AI-assisted software projects. It gives you a ready-to-use documentation skeleton, prompt files, workflow notes, conventions, and input templates.

## What the package creates

After installation, `tx-ai-agent-sdd init <path>` generates:

- `docs/` with core artifacts: `idea.md`, `vision.md`, `HLD.md`, `LLD.md`, and `tasklist.md`
- `.ai/prompts/` with prompt templates for document generation
- `.ai/workflows/` for project workflow notes
- `.ai/conventions/` for coding and delivery conventions
- `inputs/` with example inputs for idea, vision, and user story stages
- `README.spec.md` with a short workflow guide

## Installation

From PyPI:

```bash
python -m pip install tx-ai-agent-sdd
```

From source:

```bash
python -m pip install .
```

## Usage

Create a new scaffold:

```bash
tx-ai-agent-sdd init my-project
```

Show the installed CLI version:

```bash
tx-ai-agent-sdd --version
```

Copy the scaffold into an existing non-empty directory:

```bash
tx-ai-agent-sdd init my-project --force
```

Generated layout:

```text
my-project/
  .ai/
    conventions/
    prompts/
    workflows/
  docs/
    idea.md
    vision.md
    HLD.md
    LLD.md
    tasklist.md
  inputs/
    idea.input.example.md
    story.input.example.md
    vision.input.example.md
  README.spec.md
```

## Suggested workflow

1. Fill in `inputs/idea.input.example.md`.
2. Generate `docs/idea.md`.
3. Use `idea.md` to produce `docs/vision.md`.
4. Use a user story and `vision.md` to produce `docs/HLD.md`.
5. Use `HLD.md` to produce `docs/LLD.md`.
6. Use `vision.md`, `HLD.md`, `LLD.md`, workflows, and conventions to produce `docs/tasklist.md`.
7. Run subsequent iterations through the workflow protocol.

## Why it is useful

This package is a good fit if you want a repeatable way to start AI-assisted projects without rebuilding the same folder structure and document flow every time.

## Trusted Publishing

This repository includes GitHub Actions workflows for CI and PyPI publishing without long-lived PyPI tokens:

- `.github/workflows/ci.yml` runs tests on push and pull requests.
- `.github/workflows/publish.yml` builds and publishes the package through PyPI Trusted Publishing.

To enable Trusted Publishing:

1. Push this project to GitHub.
2. In PyPI project settings, add a Trusted Publisher for your GitHub repository.
3. Point PyPI to the workflow file `.github/workflows/publish.yml`.
4. Publish by creating a GitHub Release or by running the workflow manually.
