Research Ladder (right-sized depth)
A tiered approach to answering research questions with clear stop rules, evidence capture, and escalation to Tavily/Playwright only when needed.
npx skills add majiayu000/claude-skill-registry --skill research-ladder --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
## 1. Purpose / When to use - **Use this skill when** a user asks a question that requires external research and you need to pick the *right* depth (avoid over- or under-research). - **Do NOT use this skill when**: - The answer is fully contained in the repo/workspace context. - The user explicitly wants brainstorming/opinions instead of evidence. - The task is primarily implementation (code changes) and research is not a blocker. - **Inputs expected**: - User question (what decision it supports). - Context (domain, location/jurisdiction, dates/time horizon, constraints). - Constraints (time, cost, risk tolerance, required confidence, whether verbatim passages are needed). - **Outputs expected**: - A recommendation or conclusion. - Evidence summary (what sources say, and how they support key claims). - Confidence (High/Med/Low) with explicit assumptions/uncertainties. - Concrete next actions (e.g., questions to ask a contractor, fields to verify). ## 2. Research Ladder (Tiered approach with stop rules) ### Tier 0 — Quick check (native web search) **Goal:** Get a fast, minimally sufficient answer. **Approach:** - Use native web search (Bing / built-in search) with 2–4 queries. - Pre
- 1. Purpose / When to use
- 2. Research Ladder (Tiered approach with stop rules)
- Tier 0 — Quick check (native web search)
- Tier 1 — Evidence-driven synthesis (open + extract 3–5 sources)
- Tier 2 — Deep document ingestion (Tavily MCP loop)
- Tier 3 — Interactive / gated research (Playwright)
- Decision rubric (pick the right tier)
- 3. Source-quality rubric (what to prefer)
- 4. Evidence capture pattern (run-local ledger)
- 5. Output format (how to report back)
- 6. Example (neutral)
- 7. Recovery / failure modes
What does the Research Ladder (right-sized depth) skill do?
A tiered approach to answering research questions with clear stop rules, evidence capture, and escalation to Tavily/Playwright only when needed.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill research-ladder --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.