Skip to content

ericrisco/rsc-harness / skills/business-intelligence

business-intelligence

Verified

Use when a metric (revenue, MRR, margin) needs defining once in a governed semantic layer so every dashboard, report and agent returns the same number, or when an LLM must answer data questions in plain language without hallucinating SQL. NOT chart layout (that is `dashboard`), NOT which KPIs to track (that is `kpi-framework`), NOT a hand-written query (that is `sql`).

FileKindSize
SKILL.mdskill md9.8 KB
evals/README.mdmarkdown762 B
evals/cases.yamlconfig3.1 KB
references/authoring-semantic-models.mdmarkdown5.4 KB
references/wiring-agents-and-apis.mdmarkdown3.8 KB
scripts/verify.shexecutablescript4.8 KB

evals/cases.yaml

View on GitHub
skill: business-intelligence

should_trigger:
  - prompt: "Let our team ask revenue and churn questions in plain English and always get the same number back."
    why: Governed self-serve analytics in natural language — the core promise of the semantic/metrics layer.
  - prompt: "Define MRR once so every dashboard, report, and notebook computes it identically."
    why: Single shared metric definition is exactly what this skill encodes in versioned YAML.
  - prompt: "Connect our AI agent to the warehouse so it answers data questions without hallucinating SQL."
    why: Non-obvious — sounds like generic agent wiring, but the answer is to expose a semantic layer via MCP/metrics API so the agent selects governed metrics.
  - prompt: "Sales and finance report different revenue numbers and nobody can agree which is right — fix it."
    why: Non-obvious — sounds like a data bug; the real fix is one shared metric definition in the layer, not another query.
  - prompt: "Necesito una capa semántica para que el equipo consulte ventas por región en lenguaje natural."
    why: Spanish — semantic layer plus natural-language querying over governed metrics.

should_not_trigger:
  - prompt: "Design the dashboard layout and decide which charts to show on our exec view."
    route_to: dashboard
    why: This is visual presentation and panel arrangement, not defining the trusted number.
  - prompt: "Decide which 5 KPIs our SaaS should track this year and set their targets."
    route_to: kpi-framework
    why: Choosing which metrics matter and their targets is metric strategy, not technical definition.
  - prompt: "Write me a window-function query for a running total per customer."
    route_to: sql
    why: Hand-written portable query craft, not building or querying a governed metrics layer.
  - prompt: "Build a monthly board report deck that narrates the quarter's results."
    route_to: reporting
    why: Packaging results into a recurring narrative artifact, not the live ask-anything query layer.
  - prompt: "Forecast next quarter's revenue from the last three years of data."
    route_to: forecasting
    why: Projecting future values, not reporting governed historical numbers.

capability:
  - scenario: >
      Given an orders + customers + products warehouse schema, answer:
      "monthly gross revenue by product category for the last 2 quarters, EU customers only."
      Produce (a) a MetricFlow-style semantic model and (b) the metric-query spec that answers it.
    must_include:
      - Entities declared with join keys (order primary; customer and product foreign) so the layer owns the joins.
      - A measure with an explicit agg (e.g. order_amount, agg sum).
      - A metric (gross_revenue) defined over the measure, not raw SQL.
      - A time dimension (order_date) with a declared grain (day, rolled up to month).
      - The question decomposed into metric + group-by dimension (product_category) + time grain (month) + date range (last 2 quarters) + filter (region = EU) — not a hand-written SQL query.
      - A note that the agent should call the metrics API / MCP tool rather than write SQL against the warehouse.
deprecation-and-migrationaddyosmani/agent-skillsManages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another. Use when migrating a database schema in production, such as renaming or dropping a column without downtime (expand/contract). Use when deciding whether to maintain or sunset existing code.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
performance-optimizationaddyosmani/agent-skillsOptimizes application performance across frontend, backend, queries, and databases. Use when performance requirements exist, when you suspect performance regressions, when Core Web Vitals or load times need improvement, when N+1 query patterns need fixing, or when profiling reveals bottlenecks.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
agent-cost-reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
agent-cost-reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.Stars: 118kUpdated Claude CodeCodexVerifiedInstall