https://crabagent.ai/changelog/ Release notes for the crab coding agent. en https://crabagent.ai/changelog/#v0.4.0 crab-0.4.0 2026-10-22T12:00:00Z Context compaction, in two replayable stages: deterministic tool-result shedding, then an LLM-written summary if shedding was not enough. The session file is never rewritten — compaction is recorded as an event and the model's view is rebuilt by replaying it. A meter that seeds exact context occupancy and token counters at --continue, by replaying the log rather than re-estimating. The status line is correct from the first prompt. Slash-command hints that complete commands and their arguments: /model lists profiles, /skill: lists skills. https://crabagent.ai/changelog/#v0.3.0 crab-0.3.0 2026-10-21T12:00:00Z Observability end to end. The session log became format v2 — a meta row, message rows, and typed obs events (turn start/end, LLM calls, tool calls) — with canonical-JSON content digests. Four-way token usage (prompt, completion, cache read, cache write) carried from the SSE pipeline all the way to the report. crab trace <file>: renders a session's span tree, with --json. https://crabagent.ai/changelog/#v0.2.0 crab-0.2.0 2026-10-16T12:00:00Z Fullscreen TUI with an animated running indicator (spinner plus elapsed timer). /yolo and /model slash commands. STORY.md: the self-hosting arc, published while it was still short. https://crabagent.ai/changelog/#v0.1.0 crab-0.1.0 2026-10-14T12:00:00Z The agent loop, six tools (read, write, edit, bash, glob, grep), provider profiles in config.toml, and the interactive terminal UI. --print / --continue / --yolo / --provider / --system. --provider mock: a stub provider that replays canned replies, so the loop can be debugged and tested without a network call.