Recommend a price range for each quote that maximises expected margin and win odds.
Scored across 2 tools
The two tools are clearly separated: quote_pricing runs the analysis server-side, while quote_pricing_get_engine is an explicit escape hatch for oversized exports or sandbox files, with a stated size threshold that makes the choice mechanical. The only ambiguity is internal to quote_pricing, which multiplexes six distinct operations (fit_win_model, price_recommendation, explain_price, tender_bid, discount_leakage, win_loss_report) via a mode argument rather than separate tools, so an agent must reason about modes instead of tools.
Both names share the same quote_pricing prefix, which gives a clear namespace, but the pattern is not parallel: one is a bare noun phrase and the other is noun_verb_noun. Readable and non-chaotic, but not a predictable verb_noun scheme across the set.
Two tools is thin for a domain covering model fitting, recommendation, explanation, sealed-bid simulation, discount leakage and win/loss reporting, and the entire surface is compressed into one mega-tool with a mode switch. The escape-hatch tool is justified, but the split-by-size rather than by-operation bundling makes the surface feel under-decomposed rather than well-scoped.
Coverage is strong for the stated scope: model fitting, price ranges with P(win) and expected margin, explanations, Friedman/Gates tender bids, discount leakage, win/loss reporting, snapshot reuse and a large-data fallback. Gaps are minor and largely declared out of scope (catalogue-wide list pricing, demand forecasting), though there is no explicit way to log realized outcomes or compare recommendations against subsequent deals without re-supplying history.
Recommend the quote price that maximises expected margin, from your own won / lost quote history. Inputs: quote_history (quote_ref, date, segment, customer_type, product, qty, cost, list_price, quoted_price or amount, outcome won / lost / open, competitor_count; CRM / CPQ headers as exported), deal or deals (same fields, optional quoted_price to judge), quote_model_snapshot (from an earlier run), tender + tender_history (public bid results) for sealed bids, job (fit_win_model, price_recommendation, explain_price, tender_bid, discount_leakage, win_loss_report). Returns decision cards first, a price range with P(win) and expected margin at each price, explanation, Friedman / Gates tender bid, discount waterfall and leakage, win/loss report and an updated snapshot. Uses penalised logistic regression and Friedman bidding. Under 200 closed quotes gives rules of thumb only. Never uses protected attributes or competitors' private prices. Not for list price setting across a whole catalogue, demand forecasting or loan / interest pricing. Bigger exports: quote_pricing_get_engine.
Inputs: job, deal, seed, as_of, +22 more. Returns: model, alerts, status, tender, +9 more
For exports above the quote_pricing caps (over 20,000 quotes or 50,000 tender bids), or files already in your code sandbox. Returns a complete numpy-only Python script with the same engine as quote_pricing; it reads the CSV / TSV / JSON files named in the call, prints the decision cards, and writes quote_pricing_result.json and quote_model_snapshot.json. No data reaches this server on this path.
Inputs: params, deals_path, tender_path, quote_history_path, +2 more. Returns: notes, paths, script, language, +7 more
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Size an entire sports betting slate as one portfolio, using fractional Kelly and exposure limits.
Sportsbook-derived no-vig fair value with confidence, provenance, and history over REST/MCP.
Model-priced sports picks, live in-game markets and graded results: chance, fair and max price.
Independent quant layer: fee-adjusted edge, Kelly, combos, best price: Kalshi, Polymarket US, Novig
Add one secure layer between your agents and this server.