Prompt Control Plane MCP
A community-driven MCP server that acts as a control plane for AI prompts. It allows agents to classify, score, enforce policy, lock configuration, and audit prompts before they are sent to an LLM. It supports routing across multiple providers and models based on cost and quality metrics, providing a pre-flight check and optimization pipeline.
Package and access
Local package with source-attributed installation metadata. Koskamo does not execute packages during review.
Checks and signals
Each line is separate evidence, not a trust score.
- Popularity 2 GitHub stars
What you can do with it
- Score and audit prompt quality before LLM calls
- Enforce policy and lock configuration for prompts
- Route prompts across multiple LLM providers based on cost
- Optimize prompts using a full analysis pipeline
Source and maintenance
- Last repository push
- Latest release
- v5.3.4
- Maintenance checked
Packages and installation
Source-provided package data. Koskamo has not executed or live-probed this package.
claude-prompt-optimizer-mcp 2.2.2
{
"version": "2.2.2",
"transport": "stdio",
"identifier": "claude-prompt-optimizer-mcp",
"registryType": "npm",
"environmentVariables": [
{
"name": "PROMPT_OPTIMIZER_PRO",
"format": "string",
"description": "Set to 'true' to enable Pro tier via environment variable (alternative to license key)"
},
{
"name": "PROMPT_OPTIMIZER_LOG_LEVEL",
"format": "string",
"description": "Log verbosity level: debug, info, warn, error"
}
]
}Tools
Reported by the attributed source; not independently observed by Koskamo.
Maintenance history
Public check and source-refresh events. Editorial and ownership records remain private.
- · repository · passed · metadata changed · updated
- · backfill · recorded · Pre-expansion baseline; unobserved rich metadata remains unknown
Connect from your client
Each snippet configures the same source-provided package; replace placeholders before use.
Cursor
{
"mcpServers": {
"prompt-control-plane-mcp": {
"command": "npx",
"args": [
"-y",
"claude-prompt-optimizer-mcp@2.2.2"
],
"env": {
"PROMPT_OPTIMIZER_PRO": "<YOUR_PROMPT_OPTIMIZER_PRO>",
"PROMPT_OPTIMIZER_LOG_LEVEL": "<YOUR_PROMPT_OPTIMIZER_LOG_LEVEL>"
}
}
}
}Cline
{
"mcpServers": {
"prompt-control-plane-mcp": {
"command": "npx",
"args": [
"-y",
"claude-prompt-optimizer-mcp@2.2.2"
],
"env": {
"PROMPT_OPTIMIZER_PRO": "<YOUR_PROMPT_OPTIMIZER_PRO>",
"PROMPT_OPTIMIZER_LOG_LEVEL": "<YOUR_PROMPT_OPTIMIZER_LOG_LEVEL>"
}
}
}
}VS Code
{
"servers": {
"prompt-control-plane-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"claude-prompt-optimizer-mcp@2.2.2"
],
"env": {
"PROMPT_OPTIMIZER_PRO": "<YOUR_PROMPT_OPTIMIZER_PRO>",
"PROMPT_OPTIMIZER_LOG_LEVEL": "<YOUR_PROMPT_OPTIMIZER_LOG_LEVEL>"
}
}
}
}Claude Desktop
{
"mcpServers": {
"prompt-control-plane-mcp": {
"command": "npx",
"args": [
"-y",
"claude-prompt-optimizer-mcp@2.2.2"
],
"env": {
"PROMPT_OPTIMIZER_PRO": "<YOUR_PROMPT_OPTIMIZER_PRO>",
"PROMPT_OPTIMIZER_LOG_LEVEL": "<YOUR_PROMPT_OPTIMIZER_LOG_LEVEL>"
}
}
}
}Let your agent find its tools
Find servers like Prompt Control Plane MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.