# researcher

researcher is a Claude Code prompt for answering one research question about an indication. It draws on project files and external sources, checks figures, and returns cited findings.

- prompt · `type=prompt` · Claude Code, ChatGPT · ★ 654 repo stars
- Author: anthropics · official Anthropic repository
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/anthropics/life-sciences/tree/HEAD/clinical-trial-forecast/prompts/researcher.md
- Updated: 2026-11-09
- Security check: automated, 2026-11-17 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/prompts/researcher

## What it does

researcher guides an agent through a single question for a report, using the specified indication, audience, and goal. It starts with project files, then points to the clinical_trials, pubmed, biorxiv, and chembl connectors, plus WebSearch and page-fetching tools. Each factual claim needs an inline citation, and sponsor statements must remain distinct from independent evidence.

The result contains a short bottom line, confidence rating, cited key figures, sources, gaps, and follow-up questions. The `answer_markdown` section has an 800-word ceiling. Figures should include units and timepoints; unverifiable values retain an `(unverified)` label, while values marked corrected by an independent check replace the earlier ones.

PDFs are read directly with the Read tool rather than through extracted text. A single call can cover up to 20 pages. The workflow uses the named connectors, WebSearch, and fetching tools supplied through `{{fetch_tools}}`.

## Install

### Claude Code (native platform)

1. **Plugin marketplace**
In the agent chat (not a shell):
```text
/plugin marketplace add anthropics/life-sciences
```
In the agent chat (not a shell):
```text
/plugin install clinical-trial-forecast@life-sciences
```

2. **Manually: put the files in place**
- File: `.claude/commands/researcher.md`

### ChatGPT

**How to use**
- Paste the text at the start of a chat or into project instructions

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/anthropics/life-sciences/tree/HEAD/clinical-trial-forecast/prompts/researcher.md

## When to use it
- answer one research question for a report
- combine findings from project files and external sources
- attach citations to factual claims and figures
- distinguish sponsor statements from independent evidence

## Example prompts
- Research `{{indication}}` for `{{audience}}` with the goal `{{goal}}`, and cite each factual claim.
- Answer `{{question}}` for the `{{section}}` section using project files and external sources.
- Check the reported figures and mark any that cannot be verified.

## Similar
- [comprehensive-researcher](https://affpapa.org/en/nevibecode/agents/comprehensive-researcher.md) — subagent · 735 installs
- [deep-research](https://affpapa.org/en/nevibecode/skills/199-biotechnologies-deep-research.md) — skill · 10500 installs
- [source_verification](https://affpapa.org/en/nevibecode/skills/agenticorglab-source-verification.md) — skill · ★ 776 repo stars
- [endnote-mcp](https://affpapa.org/en/nevibecode/mcp/endnote-mcp.md) — MCP server · ★ 63 repo stars
- [OmniScholar](https://affpapa.org/en/nevibecode/mcp/omnischolar.md) — MCP server · ★ 8 repo stars
- [literature-search-openalex](https://affpapa.org/en/nevibecode/skills/literature-search-openalex.md) — skill · 3400 installs

From the same set: [anthropics/life-sciences](https://affpapa.org/en/nevibecode/by/anthropics/life-sciences.md) — 29 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
