Each layer does one job well; together they replace manual venue research with an auditable, fetch-verified pipeline. The request flow today, operator-driven from one machine, is: Claude Code (harness) → Kimi Relay (router) → Nebius Token Factory (host) → Kimi K3 (model), with Tavily tools and Tenki sandboxes around the edges. The rollout target (plan §6) moves the same stages into a slimmer OpenAI-compatible agent loop running inside Tenki sandboxes.
klaude command) that points Claude Code at Kimi K3 inference endpoints instead of Anthropic's API. It injects the Tavily MCP tools per-session and forwards auth to Nebius Token Factory — the bridge between the harness and the open-weight model.klaude --help as "Kimi Relay ▸ Routing Claude Code → Nebius Token Factory."nebius-kimi-k3; Kimi API docs list kimi-k3[1m].search finds neighborhoods and venue candidates; Tavily extract fetches each venue's events page to confirm it actually covers private events and isn't a 404. Every tool call is logged line-by-line.validate_config.py enforces the schema and data-honesty rules (required fields, verified links, bolded capacities, photo-caption alts). A review card flags what a human must confirm.These are not aspirational — they're checked by scripts, not by promises.
link lacks a successful fetch in evidence.jsonl is dropped by assemble_city_config.py with a nonzero exit.[UNVERIFIED] flag — never silently removed.<b> (script-checked); estimates carry a ~ and "(est)" per the skill's research rules, so nobody overbooks.build.py guards against template drift with assert. Re-running produces byte-identical output.