research-synthesizer · davila7 — Claude Code subagent: merge research findings
research-synthesizer — a subagent that consolidates output from multiple specialist researchers into one unified analysis. It detects patterns, surfaces contradictions, and structures complex data to make it clearer and more actionable.
no license 1,377 installs updated 2026-10-24
npx claude-code-templates@latest --agent deep-research-team/research-synthesizerWhat research-synthesizer does
research-synthesizer activates after several specialized research agents finish running in parallel. It reads each researcher's output, extracts key claims and supporting evidence with citations, then groups findings by theme and identifies where researchers agree, conflict, or have gaps.
The subagent relies on file read and edit tools; web search runs only to verify a specific disputed claim. Output is a markdown summary covering major themes and contradictions, plus a JSON document with full structure: themes with evidence assessment, unique insights from each researcher type, contradictions with resolution attempts, and identified knowledge gaps.
Use this when you need to merge findings from academic-researcher, web-researcher, and technical-researcher into one report without losing detail. The researcher output files must already exist in your working directory.
What it helps with
- merge findings from multiple researchers into one analysis while preserving nuance
- identify where sources contradict and find plausible explanations
- assess confidence levels in each conclusion based on evidence quality
- organize complex data into a structure that makes patterns and gaps visible
Example prompts for the agent
- All researchers are done. Synthesize everything into a report.
- Consolidate findings on WASM adoption in server runtimes from academic-researcher, web-researcher, and technical-researcher.
- Merge research outputs on LLM fine-tuning costs and flag any contradictions between sources.
How to install research-synthesizer
1claude-code-templates
npx claude-code-templates@latest --agent deep-research-team/research-synthesizer2Manually: put the files in place
.claude/agents/research-synthesizer.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.codex/agents/research-synthesizer.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.cursor/agents/research-synthesizer.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about research-synthesizer
How do I install research-synthesizer in Claude Code?
Run npx claude-code-templates@latest --agent deep-research-team/research-synthesizer in your terminal, then point your Claude Code config to .claude/agents/research-synthesizer.md.
What does research-synthesizer do?
It takes output from multiple specialized research agents, finds common themes and contradictions across sources, assesses the strength of each finding, and produces a structured report with confidence levels for every conclusion.
What researcher files do I need?
The subagent scans your directory and finds files matching patterns like *-research*, *-analysis*, or *-findings*. If no files are found, it reports the discovery failure and asks you to confirm the file locations.
Similar subagents
All subagents for AI and ML →research-synthesizer by other authors comes first.