# Research

> Investigate a question against high-trust primary sources and return cited findings without changing the project unless explicitly requested.

- Skill: `sohaibseg/research` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add sohaibseg/research`
- Raw SKILL.md: https://api.skillmd.com/api/skills/sohaibseg/research/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Research & Search
- Author: SohaibSEG (https://skillmd.com/u/sohaibseg)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/sohaibseg/research

---


Spin up a **background agent** to do the research, so you keep working while it reads.

## Mutation boundary

Return findings in chat by default. Do not write into the repository, create branches, commit, publish, or update trackers. Write a Markdown artifact only when the user explicitly requests a file; use their stated path or the current Codex task's user-facing output directory, not the project repository, unless they explicitly choose a repository path.

Its job:

1. Investigate the question against **primary sources** (official docs, source code, specs, first-party APIs), not a secondary write-up of them. Follow every claim back to the source that owns it.
2. Return a compact cited synthesis in chat.
3. If the user requested a file, write one Markdown artifact at the explicitly approved non-project path and report its location.