Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
Python agentaianthropicclaude-codecompressioncontext-engineeringcontext-windowcursorfastapilangchainllmmcpopenaiprompt-engineeringproxypythonragtoken-optimizationtokenstypescript
export {
DEFAULT_MODEL,
DEFAULT_MODELS,
buildOpencodeConfigContent,
buildOpencodeConfigContentJson,
createHeadroomProvider,
} from "./provider.js";
export type {
HeadroomModelMapping,
HeadroomProvider,
HeadroomProviderOptions,
} from "./provider.js";
export {
compressWithHeadroom,
createHeadroomRetrieveTool,
getDefaultProxyUrl,
setDefaultProxyUrl,
} from "./retrieve.js";
export type { RetrieveToolConfig } from "./retrieve.js";
export {
HEADROOM_PLUGIN_ID,
HeadroomOpenCodePlugin,
HeadroomPlugin,
default,
headroomSetup,
} from "./plugin.js";
export type { HeadroomOpenCodePluginOptions } from "./plugin.js";
export { installHeadroomTransport } from "./transport.js";