Install
$ agentstack add mcp-bug-ops-mcpls ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo 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.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
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 →About
mcpls
[](https://crates.io/crates/mcpls) [](https://docs.rs/mcpls-core) [](https://github.com/bug-ops/mcpls/actions) [](https://codecov.io/gh/bug-ops/mcpls) [](https://blog.rust-lang.org/2025/05/15/Rust-1.88.0.html) [](LICENSE-MIT)
Stop treating code as text. Give your AI agent a compiler's understanding.
mcpls is a universal bridge between AI coding assistants and language servers. It exposes the full power of LSP — type inference, cross-reference analysis, semantic navigation — through the Model Context Protocol, enabling AI agents to reason about code the way IDEs do.
Why mcpls?
AI coding assistants are remarkably capable, but they're working blind. They see code as text, not as the structured, typed, interconnected system it actually is.
mcpls changes that. By bridging MCP and LSP, it gives AI agents access to:
- Type information — Know exactly what a variable is, not what it might be
- Cross-references — Find every usage of a symbol across your entire codebase
- Semantic navigation — Jump to definitions, implementations, type declarations
- Real diagnostics — See actual compiler errors, not hallucinated ones
- Safe refactoring — Rename symbols with confidence, workspace-wide
> [!TIP] > Zero configuration for Rust projects. Just install mcpls and rust-analyzer — ready to go.
Installation
cargo install mcpls
Pre-built binaries & other methods
Download from GitHub Releases:
| Platform | Architecture | Download | |----------|--------------|----------| | Linux | x8664 | mcpls-linux-x8664.tar.gz | | Linux | x8664 (static) | mcpls-linux-x8664-musl.tar.gz | | macOS | Intel | mcpls-macos-x8664.tar.gz | | macOS | Apple Silicon | mcpls-macos-aarch64.tar.gz | | Windows | x8664 | mcpls-windows-x8664.zip |
From source:
git clone https://github.com/bug-ops/mcpls
cd mcpls
cargo install --path crates/mcpls-cli
Prerequisites (language servers)
mcpls uses graceful degradation — if one language server fails, it continues with available servers.
Rust (rust-analyzer):
rustup component add rust-analyzer
# Or: brew install rust-analyzer (macOS)
Python (pyright):
npm install -g pyright
TypeScript:
npm install -g typescript-language-server typescript
Go (gopls):
go install golang.org/x/tools/gopls@latest
> [!IMPORTANT] > At least one language server must be available.
Quick Start
1. Configure Claude Code (~/.claude/claude_desktop_config.json):
{
"mcpServers": {
"mcpls": {
"command": "mcpls",
"args": []
}
}
}
2. Experience the difference:
You: What's the return type of process_request on line 47?
Claude: [get_hover] It returns Result where:
- Response is defined in src/types.rs:23
- ApiError is an enum with variants: Network, Parse, Timeout
You: Find everywhere ApiError::Timeout is handled
Claude: [get_references] Found 4 matches:
- src/handlers/api.rs:89 — retry logic
- src/handlers/api.rs:156 — logging
- src/middleware/timeout.rs:34 — wrapper
- tests/api_tests.rs:201 — test case
MCP Tools
Code Intelligence
| Tool | What it does | |------|--------------| | get_hover | Type signatures, documentation, inferred types at any position | | get_definition | Jump to where a symbol is defined — across files, across crates | | get_references | Every usage of a symbol in your workspace | | get_completions | Context-aware suggestions that respect types and scope | | get_document_symbols | Structured outline — functions, types, constants, imports | | workspace_symbol_search | Find symbols by name across the entire workspace |
Diagnostics & Analysis
| Tool | What it does | |------|--------------| | get_diagnostics | Real compiler errors and warnings, not guesses | | get_cached_diagnostics | Fast access to push-based diagnostics from LSP server | | get_code_actions | Quick fixes, refactorings, and source actions at a position |
Refactoring & Call Hierarchy
| Tool | What it does | |------|--------------| | rename_symbol | Workspace-wide rename with full reference tracking | | format_document | Apply language-specific formatting rules | | prepare_call_hierarchy | Get callable items at a position for call hierarchy | | get_incoming_calls | Find all callers of a function (who calls this?) | | get_outgoing_calls | Find all callees of a function (what does this call?) |
Server Monitoring
| Tool | What it does | |------|--------------| | get_server_logs | Debug LSP issues with internal log messages | | get_server_messages | User-facing messages from the language server |
Configuration
Server Heuristics
mcpls uses smart heuristics to spawn only relevant language servers. Each server checks for project markers before starting.
| Language | Server | Project Markers | |----------|--------|-----------------| | Rust | rust-analyzer | Cargo.toml, rust-toolchain.toml | | Python | pyright | pyproject.toml, setup.py, requirements.txt | | TypeScript | typescript-language-server | package.json, tsconfig.json | | Go | gopls | go.mod, go.sum | | C/C++ | clangd | CMakeLists.txt, compile_commands.json, Makefile | | Zig | zls | build.zig, build.zig.zon |
> [!TIP] > Heuristics use OR logic — if ANY marker exists, the server spawns.
Custom heuristics:
[[lsp_servers]]
language_id = "rust"
command = "rust-analyzer"
[lsp_servers.heuristics]
project_markers = ["Cargo.toml", "rust-toolchain.toml", ".rust-version"]
Environment Variables
| Variable | Description | Default | |----------|-------------|---------| | MCPLS_CONFIG | Path to configuration file | Auto-detected | | MCPLS_LOG | Log level (trace, debug, info, warn, error) | info | | MCPLS_LOG_JSON | Output logs as JSON | false |
Config file locations:
| Platform | Default Location | |----------|-----------------| | Linux | ~/.config/mcpls/mcpls.toml | | macOS | ~/.config/mcpls/mcpls.toml or ~/Library/Application Support/mcpls/ | | Windows | %APPDATA%\mcpls\mcpls.toml |
Full Configuration Example
[workspace]
roots = ["/path/to/project"]
heuristics_max_depth = 10
[[lsp_servers]]
language_id = "rust"
command = "rust-analyzer"
args = []
file_patterns = ["**/*.rs"]
timeout_seconds = 30
[lsp_servers.heuristics]
project_markers = ["Cargo.toml", "rust-toolchain.toml"]
[lsp_servers.initialization_options]
cargo.features = "all"
checkOnSave.command = "clippy"
[[language_extensions]]
extensions = ["nu"]
language_id = "nushell"
See [Configuration Reference](docs/user-guide/configuration.md) for all options.
Supported Language Servers
mcpls works with any LSP 3.17 compliant server. Battle-tested with:
View supported servers
| Language | Server | Notes | |----------|--------|-------| | Rust | rust-analyzer | Zero-config, built-in support | | Python | pyright | Full type inference | | TypeScript/JS | typescript-language-server | JSX/TSX support | | Go | gopls | Modules and workspaces | | C/C++ | clangd | compile_commands.json | | Java | jdtls | Maven/Gradle projects | | Zig | zls | build.zig support | | And 24+ others | Any LSP 3.17 server | See [docs](docs/user-guide/configuration.md) |
Architecture
View architecture diagram
flowchart TB
subgraph AI["AI Agent (Claude)"]
end
subgraph mcpls["mcpls Server"]
MCP["MCP Server(rmcp)"]
Trans["Translation Layer"]
LSP["LSP ClientsManager"]
MCP --> Trans --> LSP
end
subgraph Servers["Language Servers"]
RA["rust-analyzer"]
PY["pyright"]
TS["tsserver"]
Other["..."]
end
AI |"MCP Protocol(JSON-RPC 2.0)"| mcpls
mcpls |"LSP Protocol(JSON-RPC 2.0)"| Servers
Key design decisions:
- Single binary — No Node.js, Python, or other runtime dependencies
- Async-first — Tokio-based, handles multiple LSP servers concurrently
- Memory-safe — Pure Rust, zero
unsafeblocks - Resource-bounded — Configurable limits on documents and file sizes
Documentation
- [Getting Started](docs/user-guide/getting-started.md)
- [Configuration Reference](docs/user-guide/configuration.md)
- [Tools Reference](docs/user-guide/tools-reference.md)
- [Troubleshooting](docs/user-guide/troubleshooting.md)
Development
cargo build # Build
cargo nextest run # Test
cargo run -- --log-level debug # Run locally
Requirements: Rust 1.88+ (Edition 2024)
Contributing
Contributions welcome. See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
License
Dual-licensed under [Apache 2.0](LICENSE-APACHE) or [MIT](LICENSE-MIT) at your option.
mcpls — Because AI deserves to understand code, not just read it.
Source & license
This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: bug-ops
- Source: bug-ops/mcpls
- License: Apache-2.0
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.