Background Research Agent
Delegates a research question to a background agent that saves cited findings as a Markdown file
- Repository
- mattpocock/skills
- Stars
- 309,658
- Licence
- MIT
- Files
- 2 files
What it does
When installed, the agent can hand a research question to a separate background agent with a fresh context window, so the main session keeps working while the helper reads documentation, source code and specifications. The helper is told to rely on primary sources such as official docs and first-party APIs rather than secondhand summaries, and to cite the source for each claim. The findings are written to one Markdown file in the repository, placed where existing notes already live.
Great for
- Developers who need API or library facts checked against official docs while they continue coding
- Teams that want reading and fact-gathering delegated to a background agent
- People who want research saved as a cited Markdown note inside their repo
Skip it if
- Quick factual questions that a single answer can handle, since spinning up a background agent adds overhead
- Research needing live web sources or paywalled material the agent cannot access, where citations may be incomplete
- Non-technical users without a repo to save notes into, since the output depends on repository conventions
Try asking
- “Research how the current Stripe webhook signature verification works and save the findings to the repo notes”
- “Gather the official docs on the Next.js caching APIs and write a cited summary I can review later”
- “Delegate research on the OAuth 2.0 PKCE spec to a background agent while I keep refactoring”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 2 original files, the MIT licence and an attribution note.
research.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip research.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip research.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Name the specific sources or domains you trust, such as official docs sites, so the helper starts from primary material.
- Check the cited sources in the output file before relying on claims, since the skill does not verify them independently.
- State the target folder for the Markdown file if the repo has no existing notes convention.
What's inside
- agents/openai.yaml
- SKILL.md
Redistributed unmodified from mattpocock/skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.