claude-orator — Claude Code skill for prompt review
claude-orator scores prompts across seven dimensions and returns a reworked version. Use it to shape subagent requests or review prompts across a session.
MIT ★ 78 updated 2026-03-05
/plugin marketplace add Vvkmnn/claude-emporium
/plugin install claude-orator@claude-emporiumWhat claude-orator does
The skill rates clarity, specificity, structure, context, examples, constraints, and tone on a 1–10 scale. It then applies methods such as XML tags, few-shot examples, and tool use. Run /reprompt-orator or call orator_optimize to get the breakdown and revised prompt.
Before a subagent starts, the PreToolUse(Task) hook can suggest an optimization when its prompt lacks structure. It only makes a suggestion; it does not stop the action. Scoring is deterministic, with no LLM calls or network access, and results stay in memory rather than being written to disk. Connecting the MCP server requires claude mcp add orator -- npx claude-orator-mcp.
What it helps with
- score a prompt across seven dimensions
- rewrite a request for a subagent
- spot weak points across session prompts
- add structure to an under-specified prompt
Example prompts for the agent
- Score this request and rewrite it: “Draft a database migration plan.”
- Review the subagent prompts in this session and revise the ones that lack structure.
- Rework this prompt with suitable constraints and examples.
How to install claude-orator
1Plugin marketplace
/plugin marketplace add Vvkmnn/claude-emporium/plugin install claude-orator@claude-emporiumThis installs the whole claude-orator plugin with all its skills
2Skills CLI (Vercel)
npx skills add Vvkmnn/claude-emporium --skill claude-orator -a claude-code3Manually: put the files in place
~/.claude/skills/claude-orator/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/claude-orator/
1Skills CLI (Vercel)
npx skills add Vvkmnn/claude-emporium --skill claude-orator -a codex2Manually: put the files in place
.agents/skills/claude-orator/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add Vvkmnn/claude-emporium --skill claude-orator -a cursor2Manually: put the files in place
.cursor/skills/claude-orator/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install Vvkmnn/claude-emporium plugins/claude-orator/skills/claude-orator2Manually: put the files in place
.github/skills/claude-orator/SKILL.md1Skills CLI (Vercel)
npx skills add Vvkmnn/claude-emporium --skill claude-orator -a gemini-cli2Platform CLI
gemini skills install https://github.com/Vvkmnn/claude-emporium --path plugins/claude-orator/skills/claude-oratorManually: put the files in place
.devin/skills/claude-orator/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about claude-orator
How do I install claude-orator in Claude Code?
Add the marketplace with /plugin marketplace add Vvkmnn/claude-emporium, then install the skill with /plugin install claude-orator@claude-emporium.
What does claude-orator do?
It scores prompts across seven dimensions and returns a revised version using prompt-writing techniques. Its hook can suggest a review before a subagent starts.
Which dimensions does claude-orator score?
It rates clarity, specificity, structure, context, examples, constraints, and tone, with a score from 1 to 10 for each.
Source text
SKILL.md · MIT
---
name: claude-orator
description: Prompt rhetoric coach — deterministic scoring and restructuring using Anthropic best practices
triggers: [PreToolUse]
---
# Orator Plugin
Prompt optimization. Scores prompts across 7 dimensions and restructures them using 8 Anthropic techniques. Deterministic — no LLM calls, no network, in-memory only.
## Hooks
| Hook | When | Action |
|------|------|--------|
| **PreToolUse(Task)** | Subagent prompt lacks structure | Suggests `orator_optimize` before dispatching |
Token cost: 0 on well-structured prompts (XML tags, markdown headers, action verbs). ~50-80 on vague prompts. Never blocks — suggestion only.
## Commands
| Command | Description |
|---------|-------------|
| `/reprompt-orator <prompt>` | Optimize a prompt using Anthropic best practices |
## Workflows
### Optimize (standalone)
1. `/reprompt-orator "your prompt here"` or call `orator_optimize(prompt: "...")`
2. Review score breakdown (7 dimensions, 1-10 each)
3. Use the restructured prompt with applied techniques
### Optimize (with siblings)
1. If **historian** active: `search_conversations("prompt optimization")` to find past well-scored prompts
2. `orator_optimize(prompt: "...")` — score and restructure
3. If **praetorian** active: `save_context(type: "decisions", ...)` to preserve the optimized prompt rationale
4. If **gladiator** active: `observe(summary: "xml-tags improved code prompts by +3.2")` to track what works
### Batch review
1. Review subagent prompts across a session
2. `orator_optimize` on each under-specified prompt
3. If **vigil** active: `vigil_save("before-rewrite")` before applying changes
4. Apply restructured prompts
## Sibling Synergy
| Sibling | Value | How |
|---------|-------|-----|
| **Historian** | Past well-scored prompts as examples | `search_conversations("prompt patterns")` finds effective prompts from history |
| **Praetorian** | Preserve optimization rationale | Compact optimized prompts and their scores for future reference |
| **Gladiator** | Track what techniques work best | `observe()` records which techniques improve scores most |
| **Oracle** | Find prompt engineering tools | `search("prompt patterns")` discovers relevant community tools |
| **Vigil** | Checkpoint before batch rewrites | `vigil_save()` before applying optimized prompts across files |
## MCP Tools Reference
| Tool | Purpose |
|------|---------|
| `orator_optimize` | Score prompt across 7 dimensions, apply techniques, return restructured version |
## Scoring Dimensions
Clarity · Specificity · Structure · Context · Examples · Constraints · Tone (each 1-10)
## Techniques
System prompts · XML tags · Chain of thought · Few-shot · Prefill · Long context · Extended thinking · Tool use
## Storage
In-memory only. Zero disk storage. No databases, no external services.
## Requires
```
claude mcp add orator -- npx claude-orator-mcp
```
© vvkmnn · MIT · Vvkmnn/claude-emporium