Prompt Engineer

What this skill does

• •Categorizes prompt tasks into specific engineering scopes
• •Executes an automated multi-step optimization workflow
• •Enforces strict context hygiene for prompt generation
• •Differentiates between greenfield development and maintenance
• •Standardizes interactions between related prompt ecosystems

When to use it

• ✓When an existing prompt frequently produces incorrect or inconsistent code
• ✓When drafting instructions for complex, multi-step LLM workflows
• ✓When building new prompts from scratch based on strict requirements
• ✓When managing a library of prompts that interact with each other

When not to use it

• ✕For simple, one-off questions that do not require reusable prompt logic
• ✕When you lack the time to follow a structured, multi-step refinement process

How to invoke it

Example prompts that trigger this skill:

• “Help me optimize my current system prompt.”
• “I need to fix a prompt that keeps generating broken code.”
• “Guide me through creating a new prompt for this feature.”
• “Use the prompt engineer skill to refactor my instruction set.”

Example workflow

1. Identify a prompt needing optimization.
2. Trigger the skill to initiate step 1 (triage).
3. Review the suggested scope output from the triage phase.
4. Execute step 2 using the determined scope flag.
5. Iterate through subsequent steps based on script output until complete.

Prerequisites

• –python3
• –Claude Code environment

Pitfalls & limitations

• !Attempting to manually intervene before the tool finishes its triage
• !Ignoring the scope classification, which leads to suboptimal refinement paths
• !Providing too much irrelevant context during the optimization process

FAQ

How it compares

Unlike manual prompt tweaking, this tool forces a formal triage phase and applies repeatable logic, preventing the accumulation of technical debt caused by vague or overly broad instructions.

Source & trust

From the source: “# Prompt Engineer When this skill activates, IMMEDIATELY invoke the script. The script IS the workflow. ## Invocation Start with step 1 (triage) to determine scope: <invoke working-dir=".claude/skills/scripts" cmd="python3 -m skills.prompt_engineer.optimize --step 1" /> Then continue with determined…”

# Prompt Engineer

When this skill activates, IMMEDIATELY invoke the script. The script IS the
workflow.

## Invocation

Start with step 1 (triage) to determine scope:

<invoke working-dir=".claude/skills/scripts" cmd="python3 -m skills.prompt_engineer.optimize --step 1" />

Then continue with determined scope:

<invoke working-dir=".claude/skills/scripts" cmd="python3 -m skills.prompt_engineer.optimize --step 2 --scope <scope>" />

| Argument  | Required | Description                                   |
| --------- | -------- | --------------------------------------------- |
| `--step`  | Yes      | Current step (1 = triage, 2-6 = workflow)     |
| `--scope` | For 2+   | Required for steps 2-6. Determined by step 1. |

### Scopes

- **single-prompt**: One prompt file, general optimization
- **ecosystem**: Multiple related prompts that interact
- **greenfield**: No existing prompt, designing from requirements
- **problem**: Existing prompt(s) with specific issue to fix

Do NOT analyze or explore first. Run the script and follow its output.

Quoted from solatis/claude-config for reference — see the original for the authoritative, latest version.

Recommended Rules

Recommended Workflows

Recommended MCP Servers