Skip to main content
Glama
Nimo1987

Harness Research MCP

by Nimo1987
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Remote

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
KIMI_API_KEYNoKimi LLM API key. Required if OPENROUTER_API_KEY is not set.
NCBI_API_KEYNoOptional API key for PubMed academic paper search.
BRAVE_API_KEYNoBrave search API key. Required if TAVILY_API_KEY is not set.
TUSHARE_TOKENNoOptional token for Chinese A-share financial data.
TAVILY_API_KEYNoTavily search API key. Required if BRAVE_API_KEY is not set.
OPENROUTER_API_KEYNoOpenRouter LLM API key. Required if KIMI_API_KEY is not set.

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}

Tools

Functions exposed to the LLM to take actions

NameDescription
harness_researchA

Start a deep research session: multi-source search (Tavily/Brave/arXiv/PubMed/Tushare) + CRAAP source evaluation + cross-verification → professional HTML/DOCX/PDF report.

IMPORTANT: This tool returns IMMEDIATELY with a task_id. The research runs in the background and takes ~8-12 minutes. After calling this tool, you MUST poll harness_status with the returned task_id every 30-60 seconds until status is "completed" or "failed". Do NOT wait idle — poll actively.

Workflow:

  1. Call harness_research → get task_id (returns in <1 second)

  2. Call harness_status with task_id every 30-60s to check progress

  3. When status is "completed", harness_status returns the output file paths

Driven by affordable models like Kimi K2.5 (~$0.01/run).

harness_searchA

Quick multi-source search without generating a full report. Returns structured results from Tavily, Brave, arXiv, and PubMed. Completes in seconds.

harness_statusA

Check the progress of a research task started by harness_research.

After calling harness_research, you MUST poll this tool with the returned task_id every 30-60 seconds.

  • status "running": research is in progress, keep polling

  • status "completed": research is done, output file paths are included

  • status "failed": an error occurred, error message is included

If no task_id is provided, lists all tasks.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Deploy Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4.7/5.0

Scored across 3 tools

Disambiguation5/5

Each tool has a distinct purpose: harness_research initiates a deep research session, harness_search provides quick results, and harness_status checks progress. No overlap in functionality.

Naming Consistency5/5

All tool names follow the consistent pattern 'harness_<verb>' using underscores, with predictable conventions.

Tool Count5/5

Three tools cover the essential workflow of starting research, quick searching, and status polling. The count is well-scoped for the server's purpose.

Completeness4/5

The tool set covers the core research workflow (start, search, check progress) but lacks a cancel or delete operation for tasks, which is a minor gap.

Maintenance

ActivityInactive
ResponsivenessResponsive

Resources