#319319th of 463 in Skills, plugins and promptsFor science, health and research, research and papers
deep-research-skills
Research workflow skills for Claude Code, OpenCode and Codex
- Stars
- 2.5k
- License
- MIT
- Last commit
- Aug 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
A set of skills and sub-agent definitions that run a human-in-the-loop research workflow inside Claude Code, OpenCode or Codex. /research builds an outline of items and fields, /research-deep investigates each item with parallel web-search agents, and /research-report compiles the JSON results into a markdown report. English and Chinese versions are included.
Who it is for: Claude Code, OpenCode or Codex users who need structured web research reports
Strengths
- Outline can be extended with /research-add-items and /research-add-fields before deep research starts
- Same workflow packaged for three hosts: Claude Code, OpenCode and Codex
- Deep phase runs parallel agents per item and stores results as JSON
- Final report.md is generated from the structured results
Weaknesses
- Installation is manual file copying into ~/.claude, ~/.config/opencode or ~/.codex
- Requires pyyaml and a separate web-search agent install
- OpenCode needs OPENCODE_ENABLE_EXA=1 for usable web search
- No release published; no output quality benchmarks described
What it needs
- no GPU
- Needs Claude Code, OpenCode or Codex, pyyaml, OPENCODE_ENABLE_EXA=1 (OpenCode)
- Models: Claude Code models, OpenCode (default gpt-5.4), Codex
Also in Skills, plugins and prompts
See all 463| Rank | Project | Score |
|---|---|---|
| 1 | claude-memPersistent cross-session memory plugin for Claude Code and other coding agents | 97 out of 100 |
| 2 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 97 out of 100 |
| 3 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 96 out of 100 |
| #4 | rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex | 96 out of 100 |
| #5 | cavemanShortens coding-agent replies and compresses tool output before the model reads it | 95 out of 100 |