@yeyuan98/opencode-bioresearcher-plugin

BioResearcher Plugin

OpenCode plugin that adds agents and tools for biomedical and pharmaceutical research.

Overview

No nonsense and powerful. Agents and tools that just work for your typical biomedical and pharmaceutical research needs.

Tab to activate and start asking right away:

• The Bioresearcher Agent is jack-of-all-trades.
• The BioresearcherDR Agent is a specialist for biomed/pharma deep research (DR).

Agents

BioresearcherDR

By default, it will conduct highly detailed and in-depth research, easily surpassing all general purpose LLM applications.

To reduce research depth and make research process faster, start your question with light-research:

light-research Significance of KRAS for cancer?

To skip the Question-Clarification cycle, prompt your question with no-interview:

no-interview light-research Significance of KRAS for cancer?

Tools

Table Tools

Manipulate Excel, CSV, and ODS files with precision and smart parsing for dates, numbers, and data types.

Enable LLMs to wrangle huge tables WITHOUT overwhelming model context.

What are tools for table processing?
What are column names in XXX.xlsx?
Analyze XXX.xlsx - group by Y column, and report unique values of Z column.

Calculator

Evaluate mathematical expressions with full support for brackets, powers, and scientific notation.

Make your model accurate with numbers. No more 3.11 > 3.9 nor 1+2=5.

Use the calculator tool: (3+1.5*6-1/2)^3

Blocking Timer

Pause execution for testing or pacing operations.

Respect API rate limits. No more IP bans/blocks.

Query XXX API. You MUST include a 0.5 second delay between two API calls with the blockingTimer tool.

PubMed Parser

Parse PubMed XML files to markdown or Excel format. Supports .xml and .xml.gz files.

Analyze years of publication articles in one go. Simple and powerful.

Download pubmed article data from https://ftp.ncbi.nlm.nih.gov/pubmed/updatefiles/pubmed26n1340.xml.gz and parse to Excel format.

Reference: PubMed Download Data.

OBO Parser

Parse OBO (Open Biological and Biomedical Ontology) files to CSV format. Handles Term, Typedef, and Instance frames.

Flatten complex ontologies into analysis-ready tables.

Parse go.obo file and convert to CSV using parse_obo_file tool.

Reference: OBO Format.

JSON Tools

Extract, validate, and infer JSON schemas from data.

Robust JSON handling for LLM workflows.

Extract JSON from output.md using jsonExtract tool.
Validate data.json against schema.json using jsonValidate tool.
Infer schema from sample.json using jsonInfer tool.

Database Tools

Query MySQL and MongoDB databases with read-only access.

Explore your data warehouse without leaving the conversation.

List all tables in the database.
Describe the schema for table users.
Query orders placed in the last 30 days.

Configuration via env.jsonc in working directory. Use the env-jsonc-setup skill for guided setup.

Markdown to HTML

Convert markdown to a standalone HTML file with embedded rendering — no internet required.

Generate viewable reports from markdown, right in your workflow.

Convert report.md to report.html using markdownToHtml tool.
Render my research findings as an HTML file at /tmp/findings.html.

Skills

Skills are reusable prompt templates discovered by OpenCode's built-in skill tool.

Plugin-shipped skills are automatically copied into .opencode/skills/ at plugin load time, making them discoverable alongside user-defined skills. Skills are auto-updated when the plugin version changes. To customize a plugin skill, delete its .plugin-managed marker file.

Path	Scope
.opencode/skills/	Project (includes auto-copied plugin skills)
~/.config/opencode/skills/	Global
.claude/skills/	Claude Code compatible
.agents/skills/	Agents compatible

Supplied skills

• demo-skill: showcase skill integration and resource resolution.
• python-setup-uv: setup python runtime in your working directory with uv.
• pubmed-weekly: automated download of pubmed daily update files over the past one week.
• long-table-summary: batch-process large tables using parallel subagents for summarization.
• bioresearcher-core: core patterns and utilities (retry, JSON tools, subagent waves) for skill development.
• env-jsonc-setup: guided setup for database connection configuration (db-tools).
• gromacs-guides: reusable guides for GROMACS molecular dynamics workflows.
• pzfx-io: read, parse, write, and convert GraphPad Prism PZFX files (inspect, edit, add-column, CSV/TSV/JSON round-trips).
• bioresearcher-tests: comprehensive test suite for plugin tools and skills.

Prompt the following and follow along:

Setup python uv with skill

Installation

Add the plugin to your opencode.json:

{
  "$schema": "https://opencode.ai/config.json",
  "plugin": [
    "@yeyuan98/opencode-bioresearcher-plugin",
  ],
  "mcp": {
    "biomcp": {
      "type": "local",
      "command": ["uv", "run", "--with", "biomcp-python", "biomcp", "run"],
      "enabled": true,
      "timeout": 120000
    }
  }
}

BioMCP is absolutely required and might take time to load. You will need to have uv installed on your system too. Refer to BioMCP guide for details.

Common Issues and Solutions

Issue	Solution
Agent not appearing	Check plugin in opencode.json is typed correctly
BioMCP tool failures	Make sure you have BioMCP installed and enabled. The mcp.biomcp config in opencode.json is required — without it, biomcp tools will not exist regardless of agent permissions.
biomcp tools not available to agents	Ensure the mcp section is present in your opencode.json. Agent permission config only controls access to tools that already exist — it cannot create tools that the MCP server hasn't registered.
Skills not copying after update	Clear stale plugin cache: rm -rf ~/.cache/opencode/packages/@yeyuan98/opencode-bioresearcher-plugin@latest
Research taking a long time	Check whether subagents are generating outputs; could be slow model / API throttle / overcomplicated query / etc.

License

CC BY-NC-ND 4.0