research-report

research-report — local probes, web evidence, grounded claims, one readable page

anthropic-skills:deep-research runs the web fan-out and writes notes. This skill adds what it lacks: probes on the operator's own system, advisor checkpoints, a grounding pass, a fixed report shape and the HTML render.

Before you start

The user invoked this skill by name: go on. Claude routed it: ask once, because the run spawns several agents and a workflow, then go on after a yes.

Work in .scratchpad/research/<YYYY-MM-DD>-<slug>/, the research directory. Every file this skill names goes there.

Steps

1. Probe the real system first. Read the installed source, the schema, the config, the data volumes. Save each probe as probes/<n>-<name>.md: the exact command, its output, the date. Probes are read-only; confirm isolation before a probe that writes. Cap a long run at 5 minutes and record the cap. Done when every fact the question depends on locally has a probe file.
2. Plan with the advisor. Call advisor when the session has it. Write plan.md: the question in one sentence, the facets (3 to 6, no overlap), the deliverable shape and the stop rule (every facet reported, at most one gap round).
3. Fan out to the web. Run anthropic-skills:deep-research on the question and the facets from plan.md. It writes research_notes/<title>/ and reports/<title>.md in the directory it used for research earlier in the conversation, else in the current working directory; when it is done, move both into the research directory. Its narrative report is an input to report.md, not the deliverable. Without deep-research, spawn one whetstone:whetstone-scout per facet, all in one message, and save each claim table to notes/<facet>.md.
4. Check the stop rule with the advisor. Deep-research's own gap round is the one round; here, only check that each facet has claims and record the uncovered ones as open items. Without deep-research, a facet with no claims gets one gap round.
5. Ground every load-bearing claim. List each claim the answer or a decision rests on, with its source (URL, path:line or command), about 20 to 40 claims. Run the whetstone:verify-claims workflow with args: { question, claims, batch: 5 }. For each claim that is not CONFIRMED, correct it from the source or mark it UNVERIFIED: in the report. Save the result as grounding.md.
6. Write report.md to the shape in report-core.md. Every number cites a probe, a [key] source or UNVERIFIED:.
7. Render. Run the whetstone:html-report skill on report.md to report.html, then read every figure it lists. Re-render after the last edit to report.md.
8. Check with the advisor, then account for every agent you spawned: arrived or pending.