OpenRouter Agents MCP Server
@wheattoast11
About OpenRouter Agents MCP Server
A multi-agent research MCP server + mini client adapter - orchestrates a net of async agents or streaming swarm to conduct ensemble consensus-backed research. Each task builds its own indexed pglite database on the fly in web assembly. Includes semantic + hybrid search, SQL execu
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"openrouter-deep-research-mcp": {
"command": "npx",
"args": [
"@terminals-tech/openrouter-agents",
"--stdio"
]
}
}
}Tools
42Synchronous conversational dialogue with Zero. Maintains short-term memory within the provided messages array.
Unified research tool. async=true (default) enqueues and returns {job_id}. async=false streams results synchronously like conduct_research. Example: {query: 'What is quantum computing?', costPreference: 'low', async: true}
Single entrypoint agent tool. Routes to research, follow_up, retrieve/query, or chain. Examples: {query:'AI safety'} for research, {action:'retrieve', query:'topic', k:5} for search, {chain:[{tool:'search',params:{q:'topic'}},{tool:'get_report',params:{reportId:'1'}}]} for chaining. Max depth: NaN.
Lightweight health check. Returns pong with server time; info=true adds DB/embedder/job counts.
Get job status with a compact summary by default. Use format: 'full' to include full status + events or 'events' to return only events.
Cancel a queued or running job (best-effort).
Retrieve from KB or DB. Examples: {mode:'index', query:'AI safety', k:5} or {mode:'sql', sql:'SELECT id,query FROM research_reports', explain:true}
Get current date/time. format: 'iso'|'rfc'|'epoch'.
Evaluate a simple arithmetic expression (+,-,*,/,^,(), decimals). Safe parser.
List available MCP tools with metadata. Optionally filter by query and use semantic ranking.
Semantic search over available tools. Returns ranked matches.
Execute a read-only SELECT with bound params. Example: {sql: 'SELECT * FROM research_reports WHERE id = $1', params: ['5'], explain: true}
Batch dispatch multiple research queries in a single call. Returns job IDs or waits for completion. Example: {queries: ['topic 1', 'topic 2', {query:'topic 3', costPreference:'high'}], waitForCompletion: true}
Overview
What is OpenRouter Agents MCP Server?
OpenRouter Agents MCP Server is a production-grade MCP server for multi-agent AI research and synthesis. It orchestrates parallel research agents using OpenRouter’s model API, providing planning, parallelization, and synthesis capabilities.
How to use OpenRouter Agents MCP Server?
Install via npx @terminals-tech/openrouter-agents --stdio (default STDIO transport) and set the OPENROUTER_API_KEY environment variable. Configuration can be provided via .env or .mcp.json file. For HTTP transport, add the --http flag.
Key features of OpenRouter Agents MCP Server
- Plans, parallelizes, and synthesizes multi-agent AI research.
- Embeds query vectors to route models without an LLM call.
- Persistent knowledge base with PGlite and pgvector.
- Circuit breaker for model API fault tolerance.
- Rail Protocol for bidirectional agent communication.
- Session time-travel with undo, redo, and fork.
Use cases of OpenRouter Agents MCP Server
- Decompose complex research queries into parallel sub-tasks.
- Synthesize results from multiple agents with citations.
- Persist and search research reports and knowledge graphs.
- Backend for AI assistants requiring multi-step reasoning.
- Integrate with MCP-compatible clients like Claude, Jan AI, and Continue.
FAQ from OpenRouter Agents MCP Server
What is the default transport?
STDIO is the default transport per MCP spec. Use the --http flag explicitly for HTTP mode.
Does it require an API key?
Yes, OPENROUTER_API_KEY is required. Key rotation and cooldown are supported via optional configuration variables.
Does data persist across sessions?
Yes, reports, jobs, and the knowledge graph persist by default. Set DB_AUTO_HEAL=true for in-memory mode (no persistence).
What models are supported?
Two tiers: high‑cost (e.g., Claude Sonnet 4.5, GPT‑5.2) and low‑cost (e.g., Gemini Flash, DeepSeek). Models are selected via embedding‑based routing.
Is it compliant with the MCP specification?
Yes, it complies with MCP Specification 2025‑11‑25 (Stable) and supports core features like Tools, Resources, Prompts, and SEP extensions.
Frequently asked questions
What is the default transport?
STDIO is the default transport per MCP spec. Use the `--http` flag explicitly for HTTP mode.
Does it require an API key?
Yes, `OPENROUTER_API_KEY` is required. Key rotation and cooldown are supported via optional configuration variables.
Does data persist across sessions?
Yes, reports, jobs, and the knowledge graph persist by default. Set `DB_AUTO_HEAL=true` for in-memory mode (no persistence).
What models are supported?
Two tiers: high‑cost (e.g., Claude Sonnet 4.5, GPT‑5.2) and low‑cost (e.g., Gemini Flash, DeepSeek). Models are selected via embedding‑based routing.
Is it compliant with the MCP specification?
Yes, it complies with MCP Specification 2025‑11‑25 (Stable) and supports core features like Tools, Resources, Prompts, and SEP extensions.
Basic information
More Data & Analytics MCP servers

Forency
cheevahagadogInvestigate any website's technology stack: CMS, eCommerce, frameworks, analytics, CRM, payments, and hosting behind any URL. Ask "which of these 15 prospects use HubSpot vs Salesforce?" and get a table back. Free tier:

World Monitor
Elie HabibLive global-intelligence MCP server covering news, markets, conflicts, aviation, maritime activity, energy, climate, cyber threats, supply chains, and source-grounded regional briefs.

SportsTrackLive
STL SolutionsSportsTrackLive is a remote MCP server available at https://mcp.sportstracklive.com. SportsTrackLive is a Live Tracking solution and let you record, share and relive your Tracks! SportsTrackLive provide a real-time tra

DPF
DPF AdminDPF (Data Prompt Forge) is an AI-powered platform for data ingestion, transformation and analytics using natural language. Describe your data and the shape you want and the AI infers the schema and generates the pipeline

OAIA Arena
Give any AI trading agent a public, un-cherry-pickable track record. Commit a call before it resolves, get a hash receipt, get scored against 2,048 coin-flipping monkeys on the same window.
Comments