Open-source tool
RESYNTH research pipeline
First released June 2026, last updated July 2026.
A command-line pipeline that takes several AI deep-research reports on one topic and consolidates them into a single verified master document.
It follows systematic review methods. Every claim keeps its source, conflicts are handled in the open, and completeness is checked by machine.
The pipeline has no runtime AI dependency. The thinking is supplied by whichever AI assistant drives it from outside, and every stage can be rerun and inspected from the shell with file-backed state.
How it runs
- Installs on Windows, macOS and Linux, with a guided mode for first use
- Detects Claude Code, Codex or Gemini on the machine and can wire one in to do the thinking steps
- Writes one tailored research prompt per platform, then loads and fingerprints the reports that come back
- Extracts and compares claims, then drives the consolidation into the master document
MIT licensed. Parts of it may since have been overtaken by newer tooling.
More projects
- Ask the Atlas: an interactive questionnaire over the knowledge base Interactive capstone, October 2026
- The Offerings Atlas: mapping what 166 AI platforms cover, and what they don't Knowledge base case study, October 2026
- A local gateway for headless Claude and Codex Personal tooling, October 2026
- Building a governed AI agent on an open knowledge base Agent build case study, September 2026
- Runner: Claude study game Study game, July to September 2026
- Securing Claude Cowork: controls, gaps and safer defaults Practitioner reference, Version 1.4
- Human, script or AI? Spotting AI-driven attacks in the SOC SOC advisory white paper, Version 5
- AI and MCP access governance Portfolio proof of concept, May to July 2026
- AWS security architecture case study Fictional case study, April 2026
- Incident response planning portfolio Planning portfolio, April 2026