Prompts

MCP prompts are reusable, parameterized message templates a client surfaces as slash commands. Each CostPad prompt guides the model through an existing tool sequence; arguments autocomplete from your live model.

Workflow prompts

pricing_review

Recommend what to charge for a tier: baseline economics → price sweep → risk insights. Orchestrates get_economics → run_pricing_review → insights.

• tier — target tier name (autocompletes from your tiers). Defaults to the highest-revenue paid tier.
• objective — grossProfit | totalRevenue | grossMargin | netRevAfterFees.
• rangePct — ± sweep range percent.

cost_audit

Find where margin leaks on the cost side, then surface the riskiest findings. Orchestrates run_cost_audit → insights.

• minSeverity — info | warn | critical.

compare_pricing

Diff saved scenarios side-by-side against the current baseline. Orchestrates list_scenarios → compare_pricing.

• scenarioA, scenarioB — saved scenario names (autocomplete from list_scenarios).

Ecosystem recipe prompts

Recipe prompts chain a third-party MCP's data into CostPad. They're provider-aware in their instructions but call only CostPad tools — the model reads from the provider's own MCP, then translates the result into CostPad mutations. CostPad never talks to the providers directly (see the ecosystem model and Recipes).

import-from-stripe

Pull active subscriptions from the Stripe MCP and map them into CostPad tiers + user mix. Reads stripe.list_subscriptions; writes via update_tiers / update_mix. Ends by offering pricing_review / cost_audit.

setup-from-infra

Read real usage from Supabase / Vercel / Resend MCPs and translate it into CostPad COGS. Reads supabase.get_project, vercel.get_project, resend.get_usage; writes via update_cogs. Ends by offering cost_audit / pricing_review.