AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
MCP verified MIT Self-run

Claude Prompt Editor

mcp-midwire-claude-prompt-editor · by midwire

Desktop IDE for crafting, validating, and testing Claude API prompts with best-practices linting and Claude Code MCP integration

No reviews yet
0 installs
14 views
0.0% view→install

Install

$ agentstack add mcp-midwire-claude-prompt-editor

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/mcp-midwire-claude-prompt-editor)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
4mo ago

Declared compatibility

Claude CodeClaude DesktopCursorWindsurf

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Claude Prompt Editor? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Claude Prompt Editor

A lightweight desktop IDE for crafting, validating, and testing Claude API prompts following Anthropic's best practices. Built with Tauri 2.0 (Rust + Svelte 5), distributed as a single binary for Linux and Mac.

Features

Dual-Mode Editor

  • Source mode — Monaco-based text editor with syntax highlighting for XML tags, YAML frontmatter, and {{variable}} interpolation
  • Structure mode — Block-based visual editor where each XML section (role, instructions, examples, etc.) is a collapsible, reorderable card with enable/disable toggles

Best Practices Linting

  • 10 built-in lint rules detect common prompt issues: missing role definitions, sparse examples, negative framing, over-prompting for Claude 4.6, vague instructions, and more
  • Prompt Health panel shows real-time feedback with severity levels and fix suggestions

Claude Code Integration (MCP)

  • Built-in MCP server (default port 9780) lets Claude Code load prompts directly from the editor
  • Tools: load_prompt, list_prompts, set_variable, get_prompt_health
  • No API key needed — prompts are tested through your existing Claude Code subscription

Presets & Templates

  • 13 built-in presets (roles, constraints, output formats, example skeletons)
  • 9 starter templates (blank and use-case specific: agentic, code assistant, data extraction, classification, research/RAG, conversational)
  • Custom preset creation

Version History

  • Auto-saves versions on every file save
  • Visual diff between any two versions
  • Restore previous versions, add annotations

Requirements

  • Rust (1.70+)
  • Node.js (18+)
  • pnpm (9+)
  • Linux: libgtk-3-dev, libwebkit2gtk-4.1-dev, libcairo2-dev, libpango1.0-dev, libgdk-pixbuf-2.0-dev, librsvg2-dev
  • Mac: Xcode Command Line Tools

Quick Start

git clone git@github.com:midwire/claude-prompt-editor.git
cd claude-prompt-editor
pnpm install
pnpm tauri dev

The editor opens and the MCP server starts automatically on port 9780.

Wayland

If you see a blank window or GBM buffer / Protocol error messages on Wayland, force compositing off:

WEBKIT_DISABLE_COMPOSITING_MODE=1 GDK_BACKEND=x11 pnpm tauri dev

Claude Code Integration

To use prompts from the editor in Claude Code:

# One-time setup (while the editor is running)
claude mcp add claude-prompt-editor --transport http http://localhost:9780/mcp

Then in any Claude Code session:

> Use list_prompts to see available prompts
> Load the code-review prompt and use it to review this file

Or click Copy MCP Config in the editor toolbar and paste into your Claude Code config.

Prompt File Format

Prompts are standard Markdown files with YAML frontmatter:

---
name: "Code Review Assistant"
model: claude-opus-4-6
version: 1
tags: [coding, review]
thinking:
  type: adaptive
effort: high
---

You are an expert code reviewer specializing in {{language}}.

Review the provided code for security vulnerabilities and performance issues.

def login(u, p): return db.execute(f"SELECT * FROM users WHERE u='{u}'")
CRITICAL: SQL injection vulnerability...

Environment Variables

| Variable | Default | Description | |----------|---------|-------------| | MCP_PORT | 9780 | MCP server port | | PROMPTS_DIR | ./prompts | Path to prompts directory |

License

[MIT](LICENSE)

Source & license

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

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

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.