MPS: Skill for Claude Code

.agents/skills/mps-project-management/SKILL.md

mps-project-management is a skill for Claude Code, Codex from JetBrains/MPS. It costs 149 tokens per session (2,280 once invoked), scanned A, original, Apache-2.0.

A procedure for opening, closing, or creating a JetBrains MPS project when its MCP tools cannot work because no project is open. MCP is a way for software tools to communicate with an agent.

In plain words
What is it for?
Use it to detect the MPS installation, start or reuse an IDE instance, open a project, close it, or create an empty project without relying on unavailable MCP calls.
Why use it?
It handles the common startup failure where the IDE is on its welcome screen and the normal project tools reject every request.

Skill for Claude CodeCodex ✓ vendor

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents).

This is JetBrains/MPS's own configuration. It tells Claude Code and Codex how to work on MPS itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything MPS configures →

About the project

JetBrains MPS is a development environment for creating domain-specific languages, which are programming languages designed for a particular field or task. It provides editors with features such as completion, semantic checks, and type checking, and can generate code in languages including Java and XML.

JetBrains/MPS · 1,666 stars · on GitHub · jetbrains.com

Reuse

Borrowing it

Nothing to install: this file belongs to JetBrains/MPS. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/JetBrains/MPS/master/.agents/skills/mps-project-management/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/JetBrains/MPS

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for mps-project-management

README.md
[![agentmods](https://agentmods.dev/badge/skills/jetbrains/mps/mps-project-management/github.svg)](https://agentmods.dev/skills/jetbrains/mps/mps-project-management)
Your own site
<a href="https://agentmods.dev/skills/jetbrains/mps/mps-project-management"><img src="https://agentmods.dev/badge/skills/jetbrains/mps/mps-project-management/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for mps-project-management

Your own site · 80×15
<a href="https://agentmods.dev/skills/jetbrains/mps/mps-project-management"><img src="https://agentmods.dev/badge/skills/jetbrains/mps/mps-project-management.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 149 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,280 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00149 $0.02280
Opus 5.5 $0.00060 $0.00912
Sonnet 5.5 $0.00030 $0.00456
Haiku 4.5 $0.00015 $0.00228

Measured 8d ago against content hash cfee1fc0969d, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

mps-project-management scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/new_project_migration_xml.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agents/skills/mps-project-management/SKILL.md · 83 lines

How it starts

The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Opening and closing an MPS project for MCP

Loading companion skills

Companion names in this skill are lazy dependencies: load only those relevant to the current task. If this skill came from an MCP server, use the host's skill loader to resolve the companion's unique discovered entry URI on the same host-assigned originating server. If the host has no server-backed skill loader, stop and report that limitation; do not silently fall back to a filesystem copy. If this skill came from a filesystem catalog, load the named sibling from that same catalog at <skills-root>/<skill-name>/SKILL.md, even if remote skill loaders are also available. Do not invent a tool name or server endpoint.

There is no MCP tool that can open a project. The IntelliJ MCP server rejects every tool call — including mps_mcp_list_open_projects — before dispatch unless some already-open IDE Project matches projectPath. On the welcome screen that set is empty, so MCP cannot help. The workaround is the platform CLI: pass the project directory to a second MPS process that shares the first instance's config/system directories. DirectoryLock activates the running IDE; the second process should exit in a few seconds.

Critical Directives

  • Empty Currently open projects: {"projects":[]} means the welcome screen, not a bad path. Do not retry MCP with a guessed projectPath. Open the project via CLI first.
  • Detect source-vs-standalone before choosing a command. open -a MPS.app will not talk to an MPS started as jetbrains.mps.Launcher from a checkout. Reconstructing a JVM command is the wrong move for a standalone .app / mps.sh / mps64.exe.
  • Match idea.paths.selector (and any explicit config/system dirs). MPS (2nd inst.) uses a different selector and starts a real second IDE. See references/detect-source-vs-standalone.md.
  • Do not split process command lines on spaces. Classpaths often contain IntelliJ IDEA.app or Program Files. Prefer jcmd PID VM.command_line (or /proc/PID/cmdline on Linux). When activating an already-running instance, strip -agentlib:jdwp and the IntelliJ idea_rt.jar javaagent.
  • java_command from jcmd VM.command_line is "<main-class> [args]". Use only the first token as the main class. If MPS was itself started with a project path, the field is jetbrains.mps.Launcher /path/to/previous/project; passing it whole dies with ClassNotFoundException.
  • Do not write helper scripts or jcmd dumps into the checkout. Use $TMPDIR / %TEMP% only.
  • A non-force MODAL_BLOCKED close is still pending (details.closePending: true). The close is already queued on the EDT and completes as soon as the user answers or closes the blocking dialog. Ask the user to dismiss the dialog, then check mps_mcp_list_open_projects before retrying mps_mcp_close_project. Use force=true only if the project is still open after a timed-out or cancelled close. A timed-out force close is withdrawn (closePending: false); check mps_mcp_list_open_projects.

Read the full file on GitHub · 83 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago Changed cfee1fc0969d
  2. 1mo ago First seen · 83 lines · 149 tokens per session scan A 0dc6182f7efb

Subscribe to this mod's changes

mps-project-management is a skill published in the GitHub repository JetBrains/MPS (1,666 stars, last pushed yesterday), licensed Apache-2.0. It adds 149 tokens to every session and 2,280 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-27.

Related

mps-project-management compared with skills from other repositories

mps-project-management next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
mps-project-managementJetBrains/MPSSkill1.7kyesterdayApache-2.0A149Claude Code, Codex
langiumeclipse-langium/langium-aiSkill314d agoMITA33Claude Code
lai-gen-evalseclipse-langium/langium-aiSkill314d agoMITA42Claude Code
laieclipse-langium/langium-aiSkill314d agoMITA52Claude Code
lai-gen-descriptoreclipse-langium/langium-aiSkill314d agoMITA49Claude Code
lai-gen-mcpeclipse-langium/langium-aiSkill314d agoMITA43Claude Code
lai-gen-language-skilleclipse-langium/langium-aiSkill314d agoMITA33Claude Code