Files
Battle-tested-Research-Prompts/prompts/anthropic-automated-w2s-researcher/prompt-skill-research-thinking.md
T
MerlinH 41c94b73bf Add 2 entries from recommended candidates, reject 1
- Anthropic Automated W2S Researcher: PGR 0.97 vs 0.23 human baseline;
  full prompt stack published (system prompt, skill, critic prompts)
- SPARK pathology agents: Nature Medicine, survival-validated biomarkers;
  full CrewAI agents/tasks prompts published
- OpenAI graviton amplitudes: rejected - no prompt ever released
2026-08-20 14:49:16 +00:00

2.1 KiB

name, description
name description
research-thinking Use Claude Opus 4.6 with extended thinking (max effort) + web search for hard research problems (e.g. analyzing experiment results, designing methods, deciding what to do next, doing math/theory, literature review, etc).

Research Thinking Skill

Instructions

When this skill is invoked, call the Claude Opus 4.6 API with extended thinking using Python:

python3 -c "
import os, json, urllib.request, sys

problem = '''QUERY_HERE'''

data = json.dumps({
    'model': 'claude-opus-4-6-20250310',
    'max_tokens': 16000,
    'thinking': {'type': 'adaptive'},
    'output_config': {'effort': 'max'},
    'temperature': 1,
    'messages': [{'role': 'user', 'content': problem}]
}).encode()

req = urllib.request.Request(
    'https://api.anthropic.com/v1/messages',
    data=data,
    headers={
        'Content-Type': 'application/json',
        'x-api-key': os.environ['ANTHROPIC_API_KEY'],
        'anthropic-version': '2023-06-01'
    }
)

try:
    with urllib.request.urlopen(req, timeout=1200) as resp:
        result = json.loads(resp.read())
        for block in result.get('content', []):
            if block.get('type') == 'thinking':
                print('<thinking>')
                print(block.get('thinking', ''))
                print('</thinking>')
                print()
            elif block.get('type') == 'text':
                print(block.get('text', ''))
except urllib.error.HTTPError as e:
    print(f'Error {e.code}: {e.read().decode()}', file=sys.stderr)
    sys.exit(1)
"

Replace QUERY_HERE with the actual query. For long context queries, include all relevant data in the problem string.

Requirements

  • The ANTHROPIC_API_KEY environment variable must be set
  • Python 3 must be available

Notes

  • Uses Claude Opus 4.6 (claude-opus-4-6-20250310) -- the most capable model
  • Adaptive thinking with max effort -- Claude decides how deeply to think
  • temperature must be 1 when using thinking (API requirement)
  • Max output tokens: 16000 (excluding thinking tokens)
  • Timeout: 1200s (deep thinking can take a while)

Output

Present the full response including thinking process to the user.