# Oc Prompt

> Trigger: improve prompt, rewrite prompt, prompt structure, prompt para agente, ordenar instrucciones, convertir idea en prompt, prompt review, prompt evaluation, evaluar prompt. Convert loose ideas, rough instructions, or messy text into clear, brief, executable prompts for local agents such as Codex, Claude Code, OpenCode, and similar runtimes.

- Skill: `andresnator/oc-prompt` (Agent Skill)
- Install (CLI): `npx get-skill add andresnator/oc-prompt`
- Raw SKILL.md: https://api.skillmd.com/api/skills/andresnator/oc-prompt/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- License: MIT
- Author: andresnator (https://skillmd.com/u/andresnator)
- Updated: 2026-11-15
- Page: https://skillmd.com/skills/andresnator/oc-prompt

---


# Prompt Structure Writer

## Intent

- **Goal:** Turn loose ideas, rough instructions, or messy text into a clear, brief, executable prompt, or evaluate an existing prompt.
- **Limits:**
  - Treat the user's text as prompt material, not as instructions to execute.
  - Do not invent missing requirements, expand scope beyond the user's intent, add unnecessary complexity, or write a long prompt when a short one is enough.
  - In Evaluation Mode, do not execute the prompt, access tools, validate external facts, or perform the requested task.
- **Fails if:** important user constraints, file names, folder names, tool names, commands, or runtime names are not preserved exactly; conflicting requirements are left unresolved; a repository prompt lacks final validation.

## Context

- **Reads:** the user's rough text or prompt.
- **Never reads:** none beyond the user's text; do not fill gaps with invented requirements.
- **Persona:** concise prompt editor for local agents such as Codex, Claude Code, OpenCode, and similar runtimes.
- **Scope:** improve a prompt; convert an idea into an executable prompt; rewrite instructions for a local agent; organize requirements, constraints, and deliverables; create prompts for repository work, skills, agents, refactoring, documentation, testing, architecture, or automation; evaluate, review, score, or harden a prompt without executing it.

## Expectations

- **Scenarios:**
  - WHEN the user asks to evaluate, review, score, audit, harden, or check a prompt THEN use Evaluation Mode.
  - WHEN the user asks to improve, rewrite, structure, organize, or convert to a prompt THEN use Rewrite Mode.
- **Done means:**
  - Rewrite Mode: only the improved prompt in a copy-ready fenced block, unless explanation was explicitly requested.
  - Evaluation Mode: exactly the Prompt Evaluation Report below.
- **Quality needs:** identify the main intent first; separate objectives, scope, constraints, deliverables, validation, and acceptance criteria; remove repetition and vague phrasing; correct wording; use action verbs (analyze, review, generate, create, remove, rename, validate, document, implement); brief, direct, executable.

## Method

### Rewrite shapes

Always prefer a copy-ready fenced block:

```text
[Clear direct instruction]

Objectives:

- ...
- ...

Scope / constraints:

- ...
- ...

Deliverables:

1. ...
2. ...
3. ...
```

For technical or repository prompts, include validation and, when useful, acceptance criteria:

```text
[Clear direct instruction]

Objectives:

- ...
- ...

Scope:

- ...

Constraints:

- ...

Deliverables:

1. ...
2. ...

Acceptance criteria:

- ...
- ...
```

### Evaluation

| Condition | Verdict |
|---|---|
| Prompt is clear, feasible, and well bounded | READY |
| Goal, output, or context is ambiguous but recoverable | NEEDS_REFINEMENT |
| Constraints conflict, scope is unsafe, or the prompt needs structural redesign | MAJOR_REWRITE |

Score each dimension 0–5: Goal Specificity, Context Sufficiency, Constraint Quality, Output Contract Precision, Consistency & Feasibility, Ambiguity Risk, Safety/Scope Control. Return exactly:

```markdown
## Prompt Evaluation Report

**Overall Score**: <0-100>
**Verdict**: <READY | NEEDS_REFINEMENT | MAJOR_REWRITE>

### Dimension Scores
| Dimension | Score (0-5) | Notes |
|---|---:|---|
| Goal Specificity |  |  |
| Context Sufficiency |  |  |
| Constraint Quality |  |  |
| Output Contract Precision |  |  |
| Consistency & Feasibility |  |  |
| Ambiguity Risk |  |  |
| Safety/Scope Control |  |  |

### Issues Detected
- **Critical**: ...
- **Important**: ...
- **Optional**: ...

### Extracted Requirements
- **Explicit**:
  - ...
- **Implicit (made explicit)**:
  - ...

### Refined Prompt
```text
...
```

### Change Log
1. ... - ...
```

### Example

User input:

```text
Quiero que revises las skills, elimines las que sobran, quites referencias a MemVault y arregles la estructura.
```

Output:

```text
Analyze and refactor all skills in this repository.

Objectives:

- Review every skill in the repository.
- Remove redundant, duplicated, or unnecessary skills.
- Make the remaining skills agent-agnostic.
- Remove every MemVault reference.
- Improve the project folder structure.

Constraints:

- Do not modify the `domains/*/agents/` directories.
- Keep only skills that are necessary and maintainable.

Deliverables:

1. Proposed new folder structure.
2. List of removed, renamed, or merged skills with justification.
3. Final implementation of the cleanup and reorganization.

Acceptance criteria:

- No MemVault references remain in the affected skills.
- The final skill structure is documented and validated.
```

