Skip to content
Lobstack
GatewayOne endpoint, every model, a receipt on every callConsoleWatch what ran, what it cost, and what it would have costLob BotAgents on your own machine, with a gate before anything changesNexOur own model, learning where every request is worth sending
The Gateway routing a request and returning its cost
API OverviewThe endpoint and its shapeModelsEvery model, and its rateMCP serverThe Gateway inside your editorConsoleReading your own trafficPricingRates, per model
DocumentationEverything, by sectionCompareAgainst the alternativesAgent SkillsSKILL.md files, MITConnectorsBuilt in, plus any MCP serverStatusMeasured from real trafficGitHubThe source
AboutWhat we are building, and whyEnterpriseSSO, invoicing, an SLANewsWhat changed, and whenX / TwitterLinkedInContacthello@lobstack.ai
Start free
News

What shipped, and why.

Every post here is about something that happened in the repository — a change to the Gateway, to what a plan buys, or to a claim we had been making that turned out not to be true. For the line-by-line version, the changelog is the shorter read.

Every Gateway call comes back with a receipt

Cost used to be recomputed at read time from today's price list. It is now priced when the request is served and frozen on the ledger row.

Read more
Recent

A terminal UI for the CLI, still with zero dependencies

An Agent Skill is not a connector

Plans include dollars of model spend, not messages

Comparison pages, with a build check that fails when a claim goes stale

Archive

The price rides on the frame that carries the tokens

A streamed response cannot carry cost in a header, so the last chunk before [DONE] now carries it under `x_lobstack`, beside the usual `usage` object.

The Gateway returned 400 to every OpenAI request for three months

One sampling parameter the caller never asked for, roughly 1,395 consecutive failures, and three months of blaming the provider credentials.