Agent skill · mvanhorn
pp-tablecheck
Plan Japan dining through TableCheck: discover venues, inspect course prices and conditions, check party/date availability or a bounded shortlist, and return booking URLs.
What it needs
About 4k tokens when loaded.
What this skill does
TableCheck planning Local setup Use this checkout's binary. From its directory, build and verify it, then expose it only to the current shell: Completion: the local binary runs. No authentication is needed for the verified consumer read endpoints. Workflow 1. Resolve the trip location to explicit coordinates, cuisine key, budget meaning, dates and party. Use cuisines list --query sushi when a cuisine key is uncertain. 2. Discover a bounded shortlist with venues search. Retain stable IDs, slugs and Japanese names. Completion: each candidate matches the requested filters; search availability remains a discovery hint. 3. Fetch venues get and courses list only for shortlisted venues. Use courses get for a selected course's exact fine print and conditions. Completion: currency, quoted amount, price basis, charges and relevant conditions are either sourced or explicitly unknown. 4. Use availability check for one venue or availability scan for at most five venues and fourteen days. Completion: every requested venue/date has an observation or explicit failure, with party, local time zone and freshness retained. 5. Recheck the chosen option with --refresh, then return booking-url and the material conditions. Completion: the user has the canonical link and understands that only the booking page can confirm a named course and complete the reservation. Recipes Compact shortlist Keep a bounded set of stable identities and handoff URLs. Course conditions Read exact price and fine print before comparing costs. Venue identity Resolve source details and booking mode. Availability examples Replace these example dates with the trip dates. Times are local to the venue. JSON is compact by default; diagnostics use stderr. Use dotted --select fields to reduce live output, retaining freshness when presenting availability. A dry run returns a request plan with different fields. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills mvanhorn/tablecheck