gandazgul/runwield/research

Claude Code and agent skill by gandazgul/runwield

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Investigate a question against high-trust primary sources and capture the findings as a cited Markdown note. Use when the user wants durable research, docs/API facts gathered, ecosystem comparisons, or reading legwork saved for Ideator, Planner, Architect, or future sessions.

Install

npx vskill@latest install gandazgul/runwield/research

SKILL.md

Research

Research is the durable layer above the RunWield web tools. Use web_search, web_fetch, and web_docs_search as the canonical web, URL, and docs tools, then turn the findings into a compact Markdown artifact the repo can reuse.

Use this skill when the user wants more than an answer in chat:

  • external facts, APIs, standards, or provider behavior investigated
  • library/framework docs gathered for future implementation
  • ecosystem options compared
  • source-backed context prepared for Ideator, Planner, Architect, or a future session

Do not use this skill for ordinary one-off web lookup. Use web_search, web_fetch, or web_docs_search directly for that.

Source Discipline

Prefer primary sources:

  • official documentation
  • source repositories
  • specifications and standards
  • first-party API references
  • first-party release notes, changelogs, or migration guides

Use secondary sources only to discover primary sources or to understand broader context. Do not let a secondary source own a factual claim when a primary source exists.

For library and framework APIs, use web_docs_search first. For current ecosystem facts, use web_search. For a specific URL, use web_fetch.

Process

  1. Define the research question. State the question in one sentence and the decision or artifact it should inform.
  2. Gather sources with web tools. Use web_docs_search, web_search, or web_fetch as appropriate. Prefer

official sources and keep enough source detail to cite claims.

  1. Distinguish fact from inference. Mark sourced facts separately from your synthesis or recommendation.
  2. Write one Markdown note. Save it where the repo already keeps research notes. If there is no convention, use

docs/research/<slug>.md.

  1. Keep it reusable. Link to Plans, PRDs, ADRs, or files it informs. Do not duplicate existing durable artifacts;

reference them by path.

Note Format

md
# <Research Topic>
## Question
<The research question and what decision/artifact it informs.>
## Findings
-<Sourced fact or constraint.> Source: <URL or doc path>
-<Sourced fact or constraint.> Source: <URL or doc path>
## Inference
<Your synthesis, clearly marked as inference rather than source fact.>
## Recommendation
<The practical recommendation, if the sources support one.>
## Open Questions
-<Anything the sources did not settle.>

Rules

  • Use local project language from docs/domain-language.md when naming repo concepts.
  • Respect ADRs; do not re-litigate documented decisions unless the research explicitly calls them into question.
  • Keep citations close to the claims they support.
  • Do not store secrets, private credentials, or copied proprietary content in research notes.
  • Keep quoted text short. Prefer paraphrase plus source link.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/gandazgul/runwield/research/badge)](https://verified-skill.com/skills/gandazgul/runwield/research)

Something wrong with this skill or its page? Tell us.