v0.90.0
This release is obsolete. Please use the latest release: v0.101
MCP Steroid v0.90.0 is here! This release adds Rider debugger workflow coverage, publishes first DPAIA A/B metrics, reworks the prompt platform, and improves build and integration infrastructure.
Release Date: March 2, 2026
Note: this plugin is currently tested only on macOS.
Demo: Claude Code Debugs in Rider
In this video, Claude Code drives Rider debugger workflows through MCP Steroid, including breakpoint setup, debugger control, and IDE-driven test execution.
Reporting issues
Use “Collect Logs and Diagnostics Data” action in IntelliJ to collect information about the problem. Review the log file for specific problem.
Discord Community
Message Eugene on LinkedIn for a Discord invite: linkedin.com/in/jonnyzzz
Direct invite link: discord.gg/e9qgQ7NeTC
That is the best place to discuss the plugin, ideas, features.
Download
Custom Plugin Repository (Recommended)
Add this URL in Settings -> Plugins -> ⚙️ -> Manage Plugin Repositories…:
https://mcp-steroid.jonnyzzz.com/updatePlugins.xml
The IDE will find the plugin and notify you when updates are available.
Manual Download
SHA-256: b749a8b4394c3314082be21d79ecb0ac990b64db6979e045bf7e4af5eca272da
- Download the
.zipfile above - In IntelliJ IDEA, go to Settings -> Plugins -> ⚙️ -> Install Plugin from Disk…
- Select the downloaded ZIP file
- Restart the IDE
End User License Agreement (EULA) - By downloading and using this plugin, you agree to the terms of the EULA.
Connecting to Cloud Agents
See Getting Started.
Supported AI Agents
Claude Code, Codex, Gemini, Cursor, and any other coding agent that supports the latest MCP via HTTP Streaming.
Highlights
DPAIA Arena: First Success Metrics
As of March 2, 2026, the first completed A/B comparison set shows:
- Fix rate:
4/4with MCP Steroid and4/4without MCP - Average agent time:
412swith MCP vs526swithout MCP - Net speedup: MCP was
22%faster on average (114sgain per case) - Cost: MCP was about
10%cheaper across this 4-test experiment set - Strongest gains were on complex multi-file tasks (
21-27%faster), while simple search-and-replace tasks were effectively neutral
Prompt Platform Rework
- Added a dedicated
prompts-apimodule (ArticlePart,IdeFilter,PromptsContext) - Reworked generation around article parts with computed filters
- Migrated prompt articles to the single-article model with line-2 filters and fence annotations, including multi-IDE filter support
- Replaced SkillCatalog YAML frontmatter with article properties and removed obsolete filter code paths
- Moved prompt quality gates into the prompts pipeline (
KtBlockcompilation and debugger prompt quality coverage)
Rider and Debugger Workflow Maturity
- Added Rider coverage to IntelliJ debugger support with dedicated Rider integration/debugger test flows
- Introduced RLVR-based Rider debugger prompt iteration workflows
- Added Rider playground coverage for interactive debugging sessions
- Improved debugger prompts for event-driven suspension, idempotent breakpoints, and Rider/.NET expression handling
- Added run-test-at-caret workflows and fixed action IDs/caret targeting in debugger demos
Release Notes Summary
Additional Validations
- Added article quality contract tests and stricter prompt content checks
- Updated prompt examples to compilable fences and replaced deprecated IntelliJ 2025.3 API usages
- Included Kotlin compiler version in
KtBlockcompilation cache keys to prevent stale cache reuse - Hardened release/update validation with strict release URL and asset checks
Build and Infrastructure
- Extracted
intellij-downloaderandkotlin-climodules and moved related tests to new boundaries - Fixed IDE downloader platform resolution in containerized runs (correct archive per OS)
- Improved integration-run stability via warm snapshot bootstrap hardening and test-suite parallelism controls
- Fixed ScriptExecutor timeout test behavior
- Excluded Docker CLI tests from the default fast suite
Feedback
We’re actively learning what content and features work best for the plugin. Your feedback is invaluable!
If you encounter any issues or have suggestions, please let us know in GitHub Issues! Your feedback directly shapes the development of MCP Steroid.