| name | research |
| description | External research with WebSearch + WebFetch + MCP — structured output to .meta/references/raw/ |
/research
Gather structured external research on a question and write findings to
.meta/references/raw/. Orthogonal to /brainstorm (internal ideation) and
/tech-watch (GitHub repo deep dives).
When to use: the answer depends on facts outside your training data —
recent libraries, competitor tools, emerging patterns, state-of-the-art, or
any claim that may have changed since the knowledge cutoff.
When NOT to use: you already know the answer and it is timeless — use
/brainstorm instead. Do not use /research to avoid thinking, only to
avoid guessing.
Hard rules
- No code editing. This skill only reads and writes research artifacts.
- Always write an output file to
.meta/references/raw/ — never deliver
findings only in chat. Chat disappears; files don't.
- Always cite sources with URL + date accessed. A finding without a
source is an opinion.
- Warn if fewer than 3 distinct sources — single-source findings are
fragile.
- Soft budget: recommend stopping at 8 WebFetch calls per run. User can
override.
- No credential-gated sources — public URLs only.
- Prefer MCP when available (e.g., context7 for library docs) — higher
signal than raw WebFetch.
Three steps
1. Clarify
Reformulate the question in one precise sentence: what needs to be known,
what time horizon, what depth. If the reformulation differs from the user's
phrasing, show it and confirm before starting.
2. Research
- Run WebSearch with 2-3 query variants (direct, comparative, trend form).
- Fetch 3-5 promising URLs. Prioritize official docs > peer-reviewed >
well-known blogs > forums.
- Stop when 3 strong sources agree; continue to 5 if findings diverge.
- Group findings into 2-5 themes. Identify consensus and divergence —
divergence is often the most valuable finding.
3. Write
Write to .meta/references/raw/session-YYYY-MM-DD-research-<slug>.md
following output-schema.md (same directory as this SKILL.md). Every field
in the schema is mandatory.
Then propose one next step: /brainstorm (decision needed), /spec (scope
clear), /tech-watch (specific repo surfaced), or nothing (self-contained).
Rationalizations (why you must NOT skip steps)
| Excuse | Why it's wrong |
|---|
| "I know the answer from training data" | Training has a cutoff. Research is for current facts. |
| "One source is enough" | Single-source = fragile. Warn if you can't reach 3. |
| "I'll just summarize in chat, no file needed" | Chat context disappears. The file is the deliverable. |
| "The user seems in a hurry" | A poor artifact creates false confidence. Write the file even if brief. |
