Crate every_other_token 

every-other-token

A real-time LLM token stream interceptor for token-level interaction research.

This crate sits between the caller and the model, intercepts the token stream as it arrives over SSE, applies one of five transform strategies to tokens at configurable positions, scores model confidence at each position using the OpenAI logprob API, and routes the enriched events to a terminal renderer, a zero-dependency web UI, and an optional WebSocket collaboration room.

Feature flags

Flag	Description
sqlite-log	Persist experiment runs to a local SQLite database via store::ExperimentStore.
self-tune	Enable the self-improvement telemetry bus and tuning controller.
self-modify	Enable snapshot-based parameter mutation (requires self-tune).
intelligence	Reserved namespace for future interpretability features.
evolution	Reserved namespace for future evolutionary optimisation.
helix-bridge	HTTP bridge that polls /api/stats and pushes config patches.
redis-backing	Write-through Redis persistence for agent memory and snapshots.
wasm	WASM target bindings via wasm-bindgen.

Quickstart

export OPENAI_API_KEY="sk-..."
cargo run -- "What is consciousness?" --visual
cargo run -- "What is consciousness?" --web
cargo run -- "Explain recursion" --research --runs 20 --output results.json

Modules

cli

Command-line argument definitions and helper functions.

collab

Multiplayer collaboration: room state, participant management, WebSocket handling.

config

Optional configuration file support (#16).

error

Crate-level error type for Every-Other-Token.

heatmap

Per-position token confidence heatmap exporter.

providers

Provider plugin system and SSE wire types.

render

Terminal rendering helpers extracted from the core TokenInterceptor.

replay

research

Headless research mode and batch experiment execution.

store

SQLite-backed persistence for experiment sessions and per-run metrics.

transforms

Token transform pipeline.

web

Embedded web UI server and HTTP request handling.

Structs

ResearchSession

Aggregated statistics from one or more headless inference runs.

TokenAlternative

One alternative token and its probability (for top-K logprob display).

TokenEvent

A single processed token emitted by the streaming pipeline.

TokenInterceptor

The core streaming engine that sits between the caller and the LLM.

Functions

run_research_headless

Run runs headless inference calls, collect all TokenEvents, and return an aggregated ResearchSession. Call sites must provide a constructed interceptor (no web_tx set — events are returned via the mpsc channel).