mcp-mat
Headless MCP server for Eclipse MAT that enables analyzing Java heap dumps using built-in commands, OQL queries, and predefined reports.
README
mcp-mat
Headless MCP server for Eclipse MAT using direct java -jar org.eclipse.equinox.launcher_*.jar execution.
Features
mat_healthcheck— Validate MAT launcher and Java runtime availabilitymat_parse_report— Run predefined MAT reports (leak suspects, system overview, etc.)mat_oql_query— Execute OQL queries and return inline resultsmat_run_command— Execute 56 built-in MAT analysis commands (histogram, dominator_tree, path2gc, thread_overview, etc.)mat_index_status— Check whether MAT index artifacts exist for a heap dumpmat_oql_spec— Return OQL parser guidance and supported patterns
Supported Commands (mat_run_command)
Dominator tree analysis
dominator_tree, show_dominator_tree, immediate_dominators, big_drops_in_dominator_tree
Path to GC roots
path2gc, merge_shortest_paths, gc_roots
Histogram & object listing
histogram, delta_histogram*, list_objects, group_by_value, duplicate_classes
Leak detection
leakhunter, leakhunter2, find_leaks, find_leaks2, reference_leak
Thread analysis
thread_overview, thread_details, thread_stack
Collection analysis
collection_fill_ratio, collections_grouped_by_size, array_fill_ratio, arrays_grouped_by_size, hash_entries, map_collision_ratio, extract_list_values, hash_set_values, primitive_arrays_with_a_constant_value
Reference analysis
references_statistics, weak_references_statistics, soft_references_statistics, phantom_references_statistics, finalizer_references_statistics
Finalizer analysis
finalizer_overview, finalizer_thread, finalizer_queue, finalizer_in_processing, finalizer_thread_locals
Retained set
show_retained_set, customized_retained_set
Component & top consumers
component_report, component_report_top, top_consumers, top_consumers_html, pie_biggest_objects
String & memory waste
find_strings, waste_in_char_arrays
Heap info & misc
heap_dump_overview, unreachable_objects, system_properties, class_references, comparison_report*
Eclipse/OSGi specific
bundle_registry, leaking_bundles
Export
export_hprof
* Requires a baseline heap dump (second .hprof file).
OQL mode notes
- This server normalizes client OQL input and wraps it for MAT parse-app command mode.
- You can send OQL with or without outer quotes.
- For class ranking and top consumers, prefer
mat_parse_reportwithorg.eclipse.mat.api:overviewand parseClass_Histogram*.txt. - For object inspection, use simple field-level OQL patterns (
INSTANCEOF,OBJECTS 0x...).
Environment
All environment variables are optional:
MAT_HOMEMAT_LAUNCHERJAVA_PATH(defaultjava)MAT_XMX_MB(default4096)MAT_TIMEOUT_SEC(default1800)MAT_CONFIG_DIR(default/tmp/mat-config)MAT_DATA_DIR(default/tmp/mat-workspace)MAT_DEBUG(defaultfalse)MAT_DEBUG_LOG_DIR(default/tmp/mcp-mat-logs)MAT_PRIVACY_MODE(defaultfalse)MAT_OQL_MAX_BYTES(default16384)MAT_RESULT_PREVIEW_LINES(default20)MAT_STDIO_TAIL_CHARS(default4000)
Run
npm install
npm run build
MAT_HOME=/path/to/mat node dist/src/server.js
Test
npm test
Install via npx
No build step required. Install directly from npm:
Claude Code
Add to ~/.claude/settings.json or project .claude/settings.json:
{
"mcpServers": {
"mat": {
"command": "npx",
"args": ["-y", "mcp-mat"],
"env": {
"MAT_HOME": "/Applications/MemoryAnalyzer.app/Contents/Eclipse"
}
}
}
}
Codex
codex mcp add \
--env MAT_HOME=/Applications/MemoryAnalyzer.app/Contents/Eclipse \
mat \
npx -y mcp-mat
Install in Claude Code (from source)
Add the MCP server to Claude Code settings (~/.claude/settings.json or project .claude/settings.json):
{
"mcpServers": {
"mat": {
"command": "node",
"args": ["/path/to/mcp-mat/dist/src/server.js"],
"env": {
"MAT_HOME": "/Applications/MemoryAnalyzer.app/Contents/Eclipse"
}
}
}
}
Install in Codex (from source)
Build first:
cd /path/to/mcp-mat
npm install
npm run build
Add MCP server to Codex:
codex mcp add \
--env MAT_HOME=/Applications/MemoryAnalyzer.app/Contents/Eclipse \
mat \
node /path/to/mcp-mat/dist/src/server.js
Verify:
codex mcp list
codex mcp get mat --json
If you need to update the entry:
codex mcp remove mat
# then run codex mcp add ... again
Equivalent ~/.codex/config.toml entry:
[mcp_servers.mat]
command = "node"
args = ["/path/to/mcp-mat/dist/src/server.js"]
[mcp_servers.mat.env]
MAT_HOME = "/Applications/MemoryAnalyzer.app/Contents/Eclipse"
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.