Executes OpenSearch DSL queries and returns synthesized output (table, list, summary, compact JSON) instead of raw JSON, supporting all query types including aggregations, nested queries, and has_parent queries.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@MCP Insights Proxy Show my top 10 most engaged posts from last week"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Context-efficient MCP server that executes OpenSearch queries and returns synthesized output instead of raw JSON.
Full query flexibility, synthesized output only.
You retain 100% of OpenSearch DSL capabilities. The proxy only transforms the output.
┌──────────────────────────────────────────────────────────────────────────┐
│ BEFORE (direct OpenSearch MCP) │
│ Query → OpenSearch → 500+ lines raw JSON → Context window 💥 │
├──────────────────────────────────────────────────────────────────────────┤
│ AFTER (this proxy) │
│ Query → Proxy → OpenSearch → Proxy formats → 20-50 lines → Context ✅ │
└──────────────────────────────────────────────────────────────────────────┘Related MCP server: MCP of MCPs
v1 (Limited) | v2 (Full Flexibility) |
5-6 predefined tools | 1 main tool accepting ANY DSL |
Hardcoded query patterns | You build the query |
Limited aggregations | ALL aggregations supported |
No nested/has_parent | Full query DSL support |
cd mcp-insights-proxy
npm install
npm run buildClaude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"insights-proxy": {
"command": "node",
"args": ["/absolute/path/to/mcp-insights-proxy/dist/index.js"],
"env": {
"OPENSEARCH_URL": "https://your-cluster:9200",
"OPENSEARCH_INDEX": "channel_posts",
"OPENSEARCH_USER": "username",
"OPENSEARCH_PASS": "password"
}
}
}
}opensearch_query (Main Tool)Execute any OpenSearch DSL query. Full flexibility.
Parameters:
Parameter | Type | Description |
| object | Required. Full OpenSearch DSL query object |
| string | Index name (default: |
| enum |
|
| number | Max hits to show (default: 20, max: 50) |
Example - Complex aggregation with nested sub-aggs:
{
"query": {
"size": 0,
"query": {
"bool": {
"must": [
{"term": {"join_field": "post"}},
{"term": {"channel.type": "ig"}}
],
"filter": [
{"range": {"published_at": {"gte": "now-30d"}}}
]
}
},
"aggs": {
"by_hashtag": {
"terms": {"field": "hashtags", "size": 10},
"aggs": {
"avg_engagement": {"avg": {"field": "engagement"}},
"top_creators": {"terms": {"field": "channel.name", "size": 3}}
}
}
}
}
}Example - has_parent query:
{
"query": {
"query": {
"bool": {
"must": [
{"term": {"join_field": "post"}},
{
"has_parent": {
"parent_type": "channel",
"query": {
"bool": {
"must": [
{"term": {"channel.geo.country.code": "IT"}},
{"range": {"channel.followers": {"gte": 100000}}}
]
}
}
}
}
]
}
},
"sort": [{"engagement": "desc"}],
"size": 10,
"_source": ["channel.name", "engagement", "published_at"]
}
}opensearch_countQuick count without full query overhead.
{
"query": {
"bool": {
"must": [
{"term": {"join_field": "post"}},
{"term": {"hashtags": "skincare"}}
]
}
}
}opensearch_mappingGet field list for an index.
auto (default)Automatically detects:
Aggregation-only → summary format
Few hits (≤5) → list format
Many hits → table format
table| # | name | type | engagement | likes | published_at |
|---|------|------|------------|-------|--------------|
| 1 | creator1 | IG | 156K | 142K | 2025-01-15 |
| 2 | creator2 | TT | 98K | 89K | 2025-01-18 |list**1.** @creator1 (IG) • eng: 156K • likes: 142K • views: 2.3M • 2025-01-15
**2.** @creator2 (TT) • eng: 98K • likes: 89K • views: 1.8M • 2025-01-18summaryJust counts and aggregation results, no individual hits.
compact_jsonMinimal JSON with only _source (no _id, _score, _index metadata).
**by_platform:**
• ig: 45.2K
• tt: 32.1K
• yt: 12.8K**engagement_stats:** count=89.1K avg=2.3K min=0 max=1.2M sum=205M**by_hashtag:**
• **skincare** (12.4K)
**avg_engagement:** 3.2K
**top_creators:**
• creator1: 342
• creator2: 287
• **beauty** (8.7K)
**avg_engagement:** 2.8K
...Query Type | Raw JSON | Proxy Output | Savings |
Top 10 posts | ~3000 tokens | ~300 tokens | 90% |
Aggregation (5 buckets) | ~1500 tokens | ~150 tokens | 90% |
Complex nested agg | ~5000 tokens | ~400 tokens | 92% |
Variable | Required | Default | Description |
| Yes |
| Cluster URL |
| No |
| Default index |
| No | - | Basic auth username |
| No | - | Basic auth password |
npm run dev # Development with auto-reload
npm run build # Build for production
npm start # Run production build┌─────────────┐ ┌────────────────────────────────────────────┐ ┌────────────┐
│ │ │ MCP Insights Proxy │ │ │
│ Claude │────▶│ 1. Receive DSL query (any complexity) │────▶│ OpenSearch │
│ (builds │ │ 2. Execute against cluster │ │ │
│ full │◀────│ 3. Parse response │◀────│ │
│ DSL) │ │ 4. Format: table/list/summary │ │ │
│ │ │ 5. Return compact markdown │ │ │
└─────────────┘ └────────────────────────────────────────────┘ └────────────┘
│ │
│ Returns:
│ "*45.2K hits • took 23ms*
│
│ ## Aggregations
│ **by_platform:**
│ • ig: 45.2K
│ • tt: 32.1K
│
│ ## Results
│ | # | name | engagement |..."
│
└─── ~300 tokens instead of ~3000MIT
Quick document count with optional query filter. Lighter than full query.
Inputs: index, query.
Get index mapping to understand available fields. Returns compact field list.
Inputs: index.
Execute ANY OpenSearch DSL query and return synthesized results.
You have FULL FLEXIBILITY to build any query:
Complex bool queries with must/should/filter/must_not
has_parent, has_child, nested queries
Any aggregations (terms, stats, date_histogram, scripted_metric, etc.)
Script sorting, script fields, script filters
Geo queries, range queries, wildcards, match, term, etc.
The proxy executes your query and returns COMPACT formatted output instead of raw JSON.
OUTPUT FORMATS:
auto: Detects best format (default)
table: Tabular format for multiple hits
list: One-line-per-hit format
summary: Just counts and aggregations
compact_json: Minimal JSON (only _source, no metadata)
IMPORTANT: Build the full DSL query object as you normally would.
Inputs: query (required), index, max_display, output_format.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 3 tools
Each tool has a clearly distinct purpose: count for lightweight counts, query for arbitrary DSL queries, and mapping for schema inspection. No overlap or ambiguity between them.
All tool names follow the same pattern: 'opensearch_' prefix followed by a single descriptive verb/noun (count, query, mapping). This is perfectly consistent and predictable.
With 3 tools, the set is compact and focused on core OpenSearch operations (count, query, mapping). It feels slightly minimal but appropriate for a read-only insights proxy, and each tool earns its place.
The domain is OpenSearch insights/querying, and the set covers count, query, and schema discovery, which are the essential operations. Minor gaps like direct document fetch or index listing can be handled via queries or mapping, so no critical dead ends.
Hosted MCP server for LLM cost estimation, model comparison, and budget-aware routing.
LLM cost optimization gateway, repo token-waste auditor & model arbitrage MCP server.
Single-connection meta-router indexing 500+ MCP servers & A2A tools with 85% token compression and
