A native app for macOS, Windows, and Linux. Every thread your agents have written, browsable and searchable in one place — with a chat that can read your own history back to you.
v0.11.0 · Free · macOS, Windows & Linux · all builds
Open any session as a clean, rendered transcript — user turns, assistant turns, tool calls folded away. Jump between threads without leaving the app.
A built-in, provider-agnostic chat that can search your own history and run shell commands with your approval. Bring your own key — Anthropic, OpenAI, Gemini, OpenRouter, or Ollama, or run it on your logged-in Claude Code / Codex CLI with no key at all.
Free heuristic TODO extraction, plus opt-in LLM distillation of decisions, gotchas, and summaries — with cross-thread semantic recall of past decisions and gotchas. Runs on local Ollama, your logged-in Claude Code / Codex CLI, or a cloud API key.
A synthesized, cited answer over your own threads, with [thread N] citations back to the sources it used. Needs an LLM engine — the same one that powers distillation.
Each repo gets a durable memory — its decisions, gotchas, and open TODOs aggregated across every session, with an LLM brief and a managed .callimachus/memory.md agents can read. Curate facts by hand, and let agents write back: close TODOs and record new decisions and gotchas, in the app, the CLI, or over MCP.
Send any thread to your vault as a clean note — optionally with an AI-written summary of the decisions, gotchas, and TODOs buried in it.
A dashboard of threads and messages per source and role, your busiest projects, how much of the archive is embedded for semantic search, and a spend card estimating what your AI coding cost by model.
A paginated, size-aware table to prune old threads and reclaim disk space when the archive gets heavy.
The same local index powers the CLI, the editor extension, and the MCP server. Index once; reach it from anywhere.
Every thread your agents have written, filed by source and project. Keyword fused with on-device meaning, so the half-remembered one is a keystroke away.
Each repo's decisions, gotchas, and open TODOs, distilled across every session into a durable memory your agents can actually read.
A synthesized, cited answer drawn from your own past sessions, with [thread N] references back to the source it used.
A year of work at a glance, and the mistakes you keep re-running across tools, so you fix the pattern instead of the instance.
Threads, messages, and semantic coverage at a glance, plus an estimate of what your AI coding actually cost, broken down by model and by your priciest threads.
A provider-agnostic chat that searches your own history before it answers. Bring a key, or run it keyless on your logged-in Claude Code or Codex CLI.