← ClaudeAtlas

business-researchlisted

Collect business-value data for an AI skill or workflow into a provenance-tagged run package — instrument discovery, pinned-window collection, immutable raw snapshots
tony/ai-workflow-plugins · ★ 2 · AI & Automation · score 59
Install: claude install-skill tony/ai-workflow-plugins
# Business Research Collect the data behind a business-value claim for an AI skill or agentic workflow — productivity, time saved, quality, capacity, delivery outcomes — into the interim run package the report commands render from. Value is stated in engineer-hours, cycle time, throughput, quality, and capacity. Never money. Read these first; they bind every step: - `references/interim-format.md` — the run package layout and where runs live. - `references/provenance.md` — the four tags, the anti-inflation rules, and the no-currency contract. - `references/measurement.md` — what the data must feed: the saving formulas, statistics discipline, the counterfactual ladder, and the quality guardrails. - `references/instruments.md` — per-instrument probes, collection discipline (timezone and cohort conventions, the GraphQL filteredCount trap, absence snapshots), and the collection targets in query-ready form. User arguments: $ARGUMENTS ## Context Today — run this command and read the output: ```bash date +%F ``` Kernel (WSL detection) — run this command and read the output: ```bash grep -qi microsoft /proc/version 2>/dev/null && echo "WSL" || echo "(not WSL)" ``` Instruments on PATH — run this command and read the output: ```bash for c in git gh jq; do command -v $c >/dev/null && echo "$c: available" || echo "$c: unavailable"; done ``` gh — run this command and read the output: ```bash gh auth status 2>&1 | head -3 ``` ## Procedure ### 1. Scope and orches