changelog¶

v0.18.0 (2026-01-13)¶

changes¶

• add per-chat and per-topic default agent via /agent set command #109
• add session resume shorthand for pi runner #113
• expose sender_id and raw fields on MessageRef for plugins #112

fixes¶

• recreate stale topic bindings when topic is deleted and recreated #127
• use stdout session header for pi runner #126

docs¶

• restructure docs into diataxis format and switch to zensical #121 #125

v0.17.1 (2026-01-12)¶

fixes¶

• fix telegram /new command crash #106
• track telegram sessions for plugin runs #107
• align telegram prompt upload resume flow #105

v0.17.0 (2026-01-12)¶

changes¶

• add chat session mode (session_mode = "chat") for auto-resume per chat without replying, reset with /new #102
• add message_overflow = "split" to send long responses as multiple messages instead of trimming #101
• add show_resume_line option to hide resume lines when auto-resume is available #100
• add auto_put_mode = "prompt" to start a run with the caption after uploading a file #97
• expose thread_id to plugins via run context #99
• use tomli-w for config serialization #103
• add voice_transcription_model setting for local whisper servers #98

docs¶

• document chat sessions, message overflow, and voice transcription model settings

v0.16.0 (2026-01-12)¶

fixes¶

• harden telegram file transfer handling #84

changes¶

• simplify runtime, config, and telegram internals #85
• refactor telegram boundary types #90

docs¶

• add tips section to user guide
• rework readme

v0.15.0 (2026-01-11)¶

changes¶

• add telegram file transfer support #83

docs¶

• document telegram file transfers #83

v0.14.1 (2026-01-10)¶

changes¶

• add topic scope and thread-aware replies for telegram topics #81

docs¶

• update telegram topics docs and user guide for topic scoping #81

v0.14.0 (2026-01-10)¶

changes¶

• add telegram forum topics support with /topic command for binding threads to projects/branches, persistent resume tokens per topic, and /ctx for inspecting or updating bindings #80
• add inline cancel button to progress messages #79
• add config hot-reload via watchfiles #78

docs¶

• add user guide and telegram topics documentation #80

v0.13.0 (2026-01-09)¶

changes¶

• add per-project chat routing #76

fixes¶

• hardcode codex exec flags #75
• reuse project root for current branch when resolving worktrees #77

docs¶

• normalize casing in the readme and changelog

v0.12.0 (2026-01-09)¶

changes¶

• add optional telegram voice note transcription (routes transcript like typed text) #74

fixes¶

• fix plugin allowlist matching and windows session paths #72

docs¶

• document telegram voice transcription settings #74

v0.11.0 (2026-01-08)¶

changes¶

• add entrypoint-based plugins for engines/transports plus a takopi plugins command and public API docs #71

fixes¶

• create pi sessions under the run base dir #68
• skip git repo checks for codex runs #66

v0.10.0 (2026-01-08)¶

changes¶

• add transport registry with --transport overrides and a takopi transports command #69
• migrate config loading to pydantic-settings and move telegram credentials under [transports.telegram] #65
• include project aliases in the telegram slash-command menu with validation and limits #67

fixes¶

• validate worktree roots instead of treating nested paths as worktrees #63
• harden onboarding with clearer config errors, safe backups, and refreshed command menu wording #70

docs¶

• add architecture and lifecycle diagrams
• call out the default worktrees directory #64
• document the transport registry and onboarding changes #69

v0.9.0 (2026-01-07)¶

projects and worktrees¶

• register repos with takopi init <alias> and target them via /project directives
• route runs to git worktrees with @branch — takopi resolves or creates worktrees automatically
• replies preserve context via ctx: project @branch footers, no need to repeat directives
• set default_project to skip the /project prefix entirely
• per-project default_engine and worktree_base configuration

changes¶

• transport/presenter protocols plus transport-agnostic exec_bridge
• move telegram polling + wiring into takopi.telegram with transport/presenter adapters
• list configured projects in the startup banner

fixes¶

• render ctx: footer lines consistently (backticked + hard breaks) and include them in final messages

breaking¶

• remove takopi.bridge; use takopi.runner_bridge and takopi.telegram instead

docs¶

• add a projects/worktrees guide and document takopi init behavior in the readme

v0.8.0 (2026-01-05)¶

changes¶

• queue telegram requests with rate limits and retry-after backoff #54

docs¶

• improve documentation coverage #52
• align runner guide with factory pattern
• add missing pr links in the changelog

v0.7.0 (2026-01-04)¶

changes¶

• migrate logging to structlog with structured pipelines and redaction #46
• add msgspec schemas for jsonl decoding across runners #37

v0.6.0 (2026-01-03)¶

changes¶

• interactive onboarding: run takopi to set up bot token, chat id, and default engine via guided prompts #39
• lockfile to prevent multiple takopi instances from racing the same bot token #30
• re-run onboarding anytime with takopi --onboard

v0.5.3 (2026-01-02)¶

changes¶

• default claude allowed tools to ["Bash", "Read", "Edit", "Write"] when not configured #29

v0.5.2 (2026-01-02)¶

changes¶

• show not installed agents in the startup banner (while hiding them from slash commands)

fixes¶

• treat codex reconnect notices as non-fatal progress updates instead of errors #27
• avoid crashes when codex tool/file-change events omit error fields #27

v0.5.1 (2026-01-02)¶

changes¶

• relax telegram ACL to check chat id only, enabling use in group chats and channels #26
• improve onboarding documentation and add tests #25

v0.5.0 (2026-01-02)¶

changes¶

• add an opencode runner via the opencode cli with json event parsing and resume support #22
• add a pi agent runner via the pi cli with jsonl streaming and resume support #24
• document the opencode and pi runners, event mappings, and stream capture tips

fixes¶

• fix path relativization so progress output does not strip sibling directories #23
• reduce noisy debug logging from markdown_it/httpcore

v0.4.0 (2026-01-02)¶

changes¶

• add auto-router runner selection with configurable default engine #15
• make auto-router the default entrypoint; subcommands or /{engine} prefixes override for new threads
• add /cancel + /{engine} command menu sync on startup
• show engine name in progress and final message headers
• omit progress/action log lines from final output for cleaner answers #21

fixes¶

• improve codex exec error rendering with stderr extraction #18
• preserve markdown formatting and resume footer when trimming long responses #20

v0.3.0 (2026-01-01)¶

changes¶

• add a claude code runner via the claude cli with stream-json parsing and resume support #9
• auto-discover engine backends and generate cli subcommands from the registry #12
• add BaseRunner session locking plus a JsonlSubprocessRunner helper for jsonl subprocess engines
• add jsonl stream parsing and subprocess helpers for runners
• lazily allocate per-session locks and streamline backend setup/install metadata
• improve startup message formatting and markdown rendering
• add a debug onboarding helper for setup troubleshooting

breaking¶

• runner implementations must define explicit resume parsing/formatting (no implicit standard resume pattern)

fixes¶

• stop leaking a hidden engine-id cli option on engine subcommands

docs¶

• add a runner guide plus claude code docs (runner, events, stream-json cheatsheet)
• clarify the claude runner file layout and add guidance for jsonl-based runners
• document "minimal" runner mode: started+completed only, completed-only actions allowed

v0.2.0 (2025-12-31)¶

changes¶

• introduce runner protocol for multi-engine support #7
• normalized event model (started, action, completed)
• actions with stable ids, lifecycle phases, and structured details
• engine-agnostic bridge and renderer
• add /cancel command with progress message targeting #4
• migrate async runtime from asyncio to anyio #6
• stream runner events via async iterators (natural backpressure)
• per-thread job queues with serialization for same-thread runs
• render resume as codex resume <token> command lines
• various rendering improvements including file edits

breaking¶

• require python 3.14+
• remove --profile flag; configure via [codex].profile only

fixes¶

• serialize new sessions once resume token is known
• preserve resume tokens in error renders #3
• preserve file-change paths in action events #2
• terminate codex process groups on cancel (posix)
• correct resume command matching in bridge

v0.1.0 (2025-12-29)¶

features¶

• telegram bot bridge for openai codex cli via codex exec
• stateless session resume via `codex resume <token>` lines
• real-time progress updates with ~2s throttling
• full markdown rendering with telegram entities (markdown-it-py + sulguk)
• per-session serialization to prevent race conditions
• interactive onboarding guide for first-time setup
• codex profile configuration
• automatic telegram token redaction in logs
• cli options: --debug, --final-notify, --version