{"id":"coreyhaines31-deep-research","name":"deep-research","summary":"トピックについてマルチソース、マルチステップのリサーチを求めるとき――営業コール前の競合他社調査、新しいビジネスアイデアのための市場調査、ポジショニングアングル、パートナーシップやポッドキャストゲストのデューデリジェンス、技術的な意思決定調査(どのデータベース、どの認証)、あるいは「実際に理解しなければならない...…","body":"# /deep-research — Multi-source research with archive\n\nPlans, executes, and synthesizes research from multiple sources. Archives the output so the corpus compounds.\n\n## Step 1 — Frame the question\n\nRestate the research question in one tight sentence. If ambiguous, ask the user:\n- What's the decision this research will inform?\n- What's the minimum useful answer? (Saves over-researching.)\n- Any sources to prioritize or avoid?\n\nOutput: `**Research question:** <one sentence>`\n\n## Step 2 — Plan the sources\n\nPick from this menu based on the question type. Note which sources you'll hit and why.\n\n| Source | When to use | Tool |\n|---|---|---|\n| Web search (Google) | Authoritative articles, docs, official statements | `WebSearch` |\n| `/last30days` | What people are *actually saying* right now — Reddit, X, YouTube, HN, web recency | `Skill({skill: \"last30days\", args: \"<topic>\"})` |\n| Specific URLs | When the user hands over starting URLs | `WebFetch` |\n| Browsable pages (auth-walled, JS-heavy) | Pricing pages, product tours, profiles | `agent-browser` via the `compound-engineering:agent-browser` skill |\n| Memory | Prior research / decisions / context the user already captured | grep `~/.claude/memory/` |\n| Notion | If the topic touches a known Notion workspace | Direct Notion API (key in `$NOTION_API_KEY`, see `reference_notion_api.md`) |\n| Research archive | Prior `/deep-research` runs that touched this topic | grep `${MAKERSKILLS_CONFIG:-$HOME/.config/makerskills}/deep-research/archive/` |\n\nRun discovery passes **in parallel** where possible. Sequential only when one source needs another's output (e.g., agent-browser a URL discovered by WebSearch).\n\n## Step 3 — Execute discovery\n\nRun each chosen source. For each result, capture:\n- The source (URL or system)\n- 1–3 sentence summary of what was said\n- Date / recency\n- Confidence in the source (high/medium/low)\n\nDon't synthesize yet — just collect.\n\n## Step 4 — Synthesize\n\n1. **Group findings** by theme or sub-question\n2. **Contradiction check** — flag anywhere sources disagree. Don't average them; surface the disagreement.\n3. **Confidence**: high (multiple independent sources agree), medium (one strong source or several weak), low (single anecdote or speculation)\n4. **Gaps**: what would change the answer? What's NOT in the corpus?\n\n## Step 5 — Output the brief\n\nUse this template:\n\n```markdown\n# Research: <question>\n\n**Date:** <YYYY-MM-DD>\n**Decision this informs:** <one line>\n**Confidence overall:** high / medium / low\n\n## TL;DR\n<2–4 sentences with the answer>\n\n## Key findings\n\n### 1. <Finding>\n<2–4 sentences>. Sources: [1], [3], [5]\n\n### 2. <Finding>\n...\n\n## Contradictions / uncertainty\n- <where sources disagree, with each side cited>\n\n## Gaps\n- <what's missing from the corpus>\n- <what to research next to close the gap>\n\n## Recommended next steps\n1. <action>\n2. <action>\n\n## Sources\n[1] <Title> — <URL or system> (<date>) — <confidence>\n[2] ...\n```\n\n## Step 6 — Archive\n\nArchives live in `${MAKERSKILLS_CONFIG:-$HOME/.config/makerskills}/deep-research/archive/` (create the directory if missing). Never write archives inside the skill's own folder — skill installs and upgrades re-sync from source and wipe anything saved there. **Migration:** if this skill's folder contains an old `references/research-archive/` with user entries, move those files into the archive directory first.\n\nWrite the brief to `<archive dir>/<YYYY-MM-DD>-<slug>.md` so it's grep-able forever. Slug = kebab-case of the topic.\n\nAlso append a one-line entry to `<archive dir>/INDEX.md` (create if missing):\n\n```markdown\n- 2026-06-15 — [<topic>](./<filename>.md) — <one-line TL;DR>\n```\n\n## Step 7 — Surface\n\nAfter archiving:\n- Show the full brief in chat\n- Tell the user the archive path\n- Offer: *\"Push to Notion or save to a project's docs?\"*\n\n## Composes with\n\n- `business-brainstorm` — calls this skill during the market validation step\n- `/domain` — when research includes \"is the .com available\"\n- `/last30days` — one of the data sources\n\n## Notes on quality\n\n- **Always cite.** Every claim in the brief needs a source pointer.\n- **Recency matters** — note dates on each source. For fast-moving topics (AI, startups), de-weight sources >12 months old.\n- **Don't trust a single source** for high-stakes claims. Re-search until you have at least 2 independent corroborations or surface the uncertainty.\n- **No padding.** If the answer is one paragraph, return one paragraph. The template is a maximum, not a minimum.","author":"@coreyhaines31","ownerProfile":null,"authorContacts":null,"sourceUrl":"https://github.com/coreyhaines31/makerskills/tree/main/skills/deep-research","license":"MIT","category":"document","lang":"en","tokens":1134,"stars":0,"calls30d":2,"claimed":false,"visibility":"public","origin":"crawler","version":"0.1.0","createdAt":"2026-08-22","updatedAt":"2026-08-22","files":[],"requires":{"mcp":[],"tools":[]},"safety":{"flags":[],"scannedAt":"2026-08-22","hasScripts":false,"networkEndpoints":[]}}