Every change Wellknown observed on this MCP server, newest first, with what it was before and what it became. Tool-surface changes carry the definition diff. Nothing here is edited after the fact.
Added "try_tokenzero"; changed the definition of "calculate_model_arbitrage" (5 tools before, 6 now)
⟨40 unchanged words⟩ sibling, and the effect of prompt caching.PricingThe result names the pricing table version2026-10-28.it used.
Try TokenZero (no key needed) — Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatting, edge cache, cascade advice, signed receipt) on a sample or your own prompt and model, and returns the list-price saving, the signed receipt and a copy-paste keyless curl that makes a real call on a free model. Pass {} for the default, or {"model": "gpt-6.1-sol", "system": "...", "prompt": "..."}.
Certificate recorded, valid to 2027-01-23
Authorization not required
Changed the definition of "audit_repo_token_waste", "calculate_model_arbitrage" and "optimize_prompt_payload"
No API key needed. Static-analyze a public GitHub repo (e.g. {"repo": "owner/name"}) or a set of source files for ⟨8 unchanged words⟩ on extraction tasks, hot loops, non-deterministic repeatcalls.calls, and prompt-caching mis-integration (caching gated on model names, volatile values in the system prompt, random ids in history, wrong-provider cache markers, misplaced breakpoints, cost math that ignores cached tokens). Returns findings with fixes and an estimated recoverable percentage.
Changed the definition of "calculate_model_arbitrage"
Only numbers and dates differ: this server embeds live figures in the definition, so the text moves without the contract changing.
⟨22 unchanged words⟩ the effect of prompt caching. Pricing table version2026-10-26.2026-10-27.
Unknown → Live
First tool surface recorded: 5 tools (server version 0.2.7)
https://tokenzero.sh/mcp (mcp_streamable_http) — from mcp_registry, with the record
Showing the latest 8 events. The API returns up to 500 and filters by kind: ?kind=tool_surface_changed
⟨5 unchanged words⟩ ,"type":"array"},"input_tokens":{"description":"per request; default 2000","type":"number"},"model":{"description":"model id, e.g. gpt-6.1-sol or claude-sonnet-5-5","type":"string"},"output_tokens":{"description":"per request; default 400","type":"number"},"requests_per_month":{"type" ⟨14 unchanged words⟩ type":"number"}},"required":["model","input_tokens","output_tokens"],"type":"object"}
No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request ⟨34 unchanged words⟩