# Directory To Spec

> Create a spec directory with specification files per major feature for code in the current directory and its subdirectories.

- **Type:** Skill
- **Install:** `agentstack add skill-tomzx-agents-directory-to-spec`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [tomzx](https://agentstack.voostack.com/s/tomzx)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [tomzx](https://github.com/tomzx)
- **Source:** https://github.com/tomzx/agents/tree/main/skills/directory-to-spec

## Install

```sh
agentstack add skill-tomzx-agents-directory-to-spec
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Convert Directory to Spec

Analyzes code in the current directory and generates a `spec/` directory containing one specification file per major feature, each with MUST/SHOULD/MAY requirements.

## Prerequisites

- Read access to the current directory and its subdirectories
- A codebase with identifiable features or modules

## Steps

1. Read and analyze the code in the current directory and subdirectories to identify major features.
2. Create the `spec/` directory.
3. For each major feature, create a descriptively named specification file (e.g., `authentication.md`).
4. Each file must include a `requirements` section using RFC 2119 language (`MUST`, `SHOULD`, `MAY`).
5. Specifications must remain technology-agnostic: no specific library, framework, or technology choices should appear in the requirements, unless used as illustrative examples (e.g., "the system MUST provide a key-value cache, such as Redis").

**Naming rules:**
- Use descriptive, lowercase names: `authentication.md`
- Do NOT number files: ~~`01-authentication.md`~~
- Do NOT suffix with `-spec`: ~~`authentication-spec.md`~~

## Example Usage

**Scenario 1: REST API project**
Working directory contains `auth/`, `orders/`, `payments/`. Produces:
- `spec/authentication.md`
- `spec/order-management.md`
- `spec/payment-processing.md`

**Scenario 2: Single-module library**
Working directory contains a single Python library with multiple classes. Produces one spec file per major class or responsibility area.

**Scenario 3: Frontend application**
Working directory contains React components. Produces spec files organized around user-facing features rather than individual component names.

## Useful Commands Reference

| Command | Description |
|---|---|
| `mkdir spec` | Create the spec output directory |

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [tomzx](https://github.com/tomzx)
- **Source:** [tomzx/agents](https://github.com/tomzx/agents)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-tomzx-agents-directory-to-spec
- Seller: https://agentstack.voostack.com/s/tomzx
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.