AGENT SKILL

yun-research

Investigate a question against primary sources and leave a cited Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, a claim verified, or reading legwork delegated while they keep working.

License 许可
MIT
最近更新
2026-11-13
Stars
★ 1
Forks
0

01 前言 Front Matter

name
yun-research
description
Investigate a question against primary sources and leave a cited Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, a claim verified, or reading legwork delegated while they keep working.

03 SKILL.md 正文

yun-research

Research is legwork you delegate, not thinking you outsource: a background agent does the reading, you keep working, and you get back a cited document to react to.

Steps

  1. Recall first. Follow skills/_shared/memory-convention.md to recall prior research on this question, scoped to the target project. If an existing digest already answers it, surface that and stop — no agent needed.

  2. Spin up a background agent to do the reading, so the user is not blocked. Hand it the question and this brief:

    • Work from primary sources — official docs, source code, specs, first-party APIs — never a secondary write-up of them. Follow every claim back to the source that owns it.
    • Write the findings to a single Markdown file in the repo, each claim cited to its source. Save it where the repo already keeps such notes; match the existing convention, and if there is none, pick somewhere sensible and say where.
    • Read the harness (skills/, AGENTS.md) as of origin/main — git show origin/main:<path> — whenever this checkout is behind it.

    Done when the agent has returned and the file exists with a source on every claim.

  3. Persist the digest. Once the file lands, follow skills/_shared/memory-convention.md to persist a short digest — the question and its findings in a few lines — scoped to the target project. The full file stays in the repo; the digest is what downstream skills recall without re-reading it.

  4. Hand it back. Surface the file to the user. It is something to react to — the input that feeds grilling, planning, or a handoff — not a final answer.

Attribution

Adapted for this workspace from Matt Pocock's research (github.com/mattpocock/skills, MIT). Changes: renamed to the yun-<verb>-<noun> harness convention; added a recall-first step and a project-scoped digest via skills/_shared/memory-convention.md (the full cited file still lives in the repo); dropped the OpenAI agent wiring.

04 同仓库更多 Skills

matiinchauspe
matiinchauspe/yunque

yun-write-wizard

Generate an interactive bash wizard that walks a human through steps only they can perform. Use when provisioning infrastructure, setting up credentials or CI secrets, walking an unfamiliar third-party dashboard, or running a one-off migration or cutover. Don't invoke this for steps the agent can perform itself.

★ 1
matiinchauspe
matiinchauspe/yunque

yun-write-spec

Write the spec — consolidate a large task's settled decisions into one durable document before it is sliced into tickets. Use when the decisions are settled and the user wants them written up, or says "write the spec" / "escribí el spec". Declines small fixes that need no spec.

★ 1
matiinchauspe
matiinchauspe/yunque

yun-write-pr

Write a pull request title and body that show the change — its ticket, its shape, the evidence it works, and how dangerous it is to merge.

★ 1
matiinchauspe
matiinchauspe/yunque

yun-write-handoff

Compact the current session into a disposable handoff baton for the next agent, pointing at durable memory instead of duplicating it.

★ 1
matiinchauspe
matiinchauspe/yunque

yun-write-agent-docs

Use when authoring or editing a document an agent reads in this workspace — a skill, a `skills/_shared/` contract, `AGENTS.md` or `skills/README.md` — deciding how it is invoked, what lives in it versus a linked file, naming it, or diagnosing why a skill misfires, bloats, or fires unreliably.

★ 1
matiinchauspe
matiinchauspe/yunque

yun-tdd

Test-driven development — the red → green loop, tests written only at pre-agreed seams. Use when the user wants to build a feature or fix a bug test-first, or mentions TDD or red-green. Reached by the implement flow to drive a build at a seam.

★ 1