Files
codecraft/CHANGELOG.md
T
2026-11-16 13:21:35 +01:00

758 lines
38 KiB
Markdown
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# Changelog
All notable changes to CodeCraft. Versions follow [Semantic Versioning](https://semver.org/);
until 1.0, minor versions may contain breaking changes.
## [v202611-00050] – 2026-11-16
### Fixed
- The factory's planning chats (spec chat, "Plan a feature", "Cut a milestone") no longer run the review loop
of the agent mode (the "refine" rounds of build mode), which checked their answers as if they were code.
## [v202611-00049] – 2026-11-16
### Added
- **Factory versions and tags.** Every merged factory issue (merged through the factory or on the forge)
gets a version: the patch number goes up with each issue (0.1.0 → 0.1.1); the issue that completes a
milestone (the last open `M1.x` issue) raises the minor number and restarts the patch (0.0.7 → 0.1.0).
The factory commits a `VERSION` file on the default branch, tags the commit `vX.Y.Z`, closes the issue
and notes the version in it. The version shows on the repository tile. Versioning starts with the next
merged issue; issues that were done before are not counted.
- **Factory release (0.x → 1.0.0 and on).** The new "Release" button makes a release you choose the
version of: `VERSION`, tag, the Docker image for amd64 and arm64 (buildx, pushed to the registry of the
image name), the source zip and further files, and the release on the forge (Gitea and GitHub attach the
files; GitLab gets the release without attachments). How to build is described in
`.factory/release.json` in the repository (`image`, `platforms`, `dockerfile`, `context`,
`build_args`, `commands`, `files`, `name`); without it only tag, release and source zip are made. The
dialog shows the recipe first, because its commands run on the server outside the sandbox.
- The image now contains the buildx plugin.
- **The factory's checkout of the default branch is pulled** (fast-forward) after every fetch, and right after
a merge is detected and the version commit is pushed – before, the review read the files of the first clone.
Local changes or a diverged branch are left alone with a warning.
- **Settings for repetitions** per repository: *Rework rounds* (how often a failed acceptance sends the
issue back to the build, default 2, 0 = never) and *Tries* per phase (how often a phase is repeated after
a technical failure such as a timeout, default 2).
## [v202611-00048] – 2026-11-16
### Fixed
- The factory page showed a blank page since v202611-00047 (the new pull request panel declared its state
after an early return, which breaks React's hook order). The hooks now sit above the early returns.
## [v202611-00047] – 2026-11-16
### Added
- **Factory: show and approve pull requests.** Selecting an issue shows its open pull request (state,
conflicts, files, +/− lines, link) with "Approve and merge" (merge commit, squash or rebase). Merging only
happens on your click after a confirmation; the factory itself still never merges.
- **Factory: plan with a model and file the issues.** "Plan a feature" starts a chat that works out a new
feature with you; "Cut a milestone" lets a model cut a milestone of the spec into numbered deliverables
("M2.1 – …"). The chat reads the repository and ends with an issue list; a dialog lets you tick and edit
the issues, which are then created on the forge with `Spec:` and `Scope:` lines and the prerequisites.
They get the factory label only if you ask for it, so you can start them one after the other.
### Fixed
- An accepted rework build now gets a new pull request when the earlier one was merged or closed already;
before, the commits of the rework rounds stayed on the branch without any pull request.
## [v202611-00046] – 2026-11-16
### Changed
- **Factory: the pull request is opened by the acceptance**, not by the build. A build only pushes its
branch `factory/issue-<n>`; once the acceptance passes, the pull request is created with the build
report and the result table (an existing one is kept).
- The factory has its own icon (a factory instead of a toolbox).
- All factory chats (spec chats, review, build, acceptance) are collected in one sidebar folder
"Factory"; existing ones are moved there on start.
### Fixed
- Factory acceptance: scenarios outside the scope of the issue (stored scenarios that cover a whole
multi-milestone spec) get the result `out_of_scope` and no longer fail the issue. The acceptance
receives the build report to tell what the build claims to cover, and an acceptance in which no
scenario of the scope could be observed working does not pass.
## [v202611-00045] – 2026-11-16
### Fixed
- Factory acceptance: when the built branch has no `.factory/setup.json` (it was added to the default
branch later), the one from the default branch is used.
- Factory acceptance judged things outside the issue (e.g. later milestones of a large spec) and failed
the issue for them. Now only the listed scenarios count: the verdict follows the scenario results
(one failed scenario fails, unverified or out-of-scope gaps do not), and the prompts of review, build
and acceptance restrict the work to the issue's scope. An issue can name it with a `Scope: <name>`
line (e.g. `Scope: M1`); without one the review picks the first milestone that is not built yet.
## [v202611-00044] – 2026-11-16
### Added
- The factory spec chat now also writes `setup.json` (how to install and start the app). Creating the
project files it as `.factory/setup.json` in the repository, where the acceptance phase reads it; the
create dialog warns when it is missing.
## [v202611-00043] – 2026-11-16
### Added
- **Factory acceptance phase**: a finished build is checked by a different model. It starts the built
branch (as `.factory/setup.json` describes: `install`, `start`, `url`), goes through the acceptance
scenarios from the review – kept away from the builder – and uses the browser tool where it can.
Every scenario ends as pass, fail or *unverified*; unverified ones are reported openly and never
count as passed. One failed scenario sends the issue back to the build with the findings (at most
2 rework rounds, then `failed`); otherwise the issue is `done` and the pull request waits for you
to merge – the factory never merges.
- **Restart an issue** in the factory page (select an issue → "Restart at": review, build or
acceptance). Works for failed, rejected, done and waiting issues; restarting at review forgets the
stored scenarios, and restarts reset the retry counters.
### Fixed
- Factory runs interrupted by a server restart continue where they were (builds in a rework round
resume as rework).
## [v202611-00042] – 2026-11-16
### Added
- **Factory build phase**: approved issues are built automatically. The build agent works in its own git
worktree on branch `factory/issue-<n>` (inside the factory area, never in the workspace), with the
model chosen for the build phase. The server – not the agent – commits, pushes and opens the pull
request; the spec file is restored if the agent touched it. A build that makes no changes, times out or
returns no result is retried once, then marked `failed`; a `blocked` answer is reported on the issue.
Finished builds move to `factory:qa` (acceptance follows in a later version).
## [v202611-00041] – 2026-11-15
### Added
- **Software factory** (sidebar → Factory): specs arrive as issues (label `factory`) on Gitea, GitHub or
GitLab and are worked off by agents in phases, each phase with its own model, agent, permission mode
and time limit. This version contains the **review phase**: an agent reads the spec and the code and
decides *approved*, *questions* (posted as an issue comment; your answer restarts the review) or
*rejected*, and writes acceptance scenarios that stay in the database, away from the builder. Build and
acceptance follow.
- **Connections per user**: every user sets up their own accounts (Gitea, GitHub, GitLab) with their own
token; repositories use the token of their connection or one of their own. Models are free per user,
project and phase.
- **New project with a model**: *Factory → New project* opens a spec chat. When the spec is done, the
*Create project* bar picks name, connection and owner, creates the repository (private), commits the
spec as `docs/specs/<name>.md` and opens the issue that starts the workflow.
- **Overview**: board with all issues by state, a tile per project, and per run the model, status,
duration, tokens and cost, with a link into the chat of that run.
- **Separate factory area**: repositories are cloned into their own mount `/factory` (`FACTORY_DIR` in
`.env`), not into the workspace. The API token is handed to git only as a header and never reaches
agents.
## [v202611-00040] – 2026-11-15
### Fixed
- **Images you upload reach Claude Code**: its chats read the history without the image bytes, so the
files it was pointed to were empty. Images without data are no longer written as empty files.
## [v202611-00039] – 2026-11-15
### Changed
- A chat's colour no longer tints the chat background (pane, message field, agent bubbles, top bar
fill); it still marks the top bar line, the list entry, the badge and your own bubbles.
### Fixed
- **Claude Code can ask questions again**: its `AskUserQuestion` was always rejected; now the question
appears as a question card in the chat (single and multiple choice) and the answers go back to it.
## [v202611-00038] – 2026-11-15
### Changed
- The chat menu in the sidebar has a **Delete** entry (with the question about the chat's own folder);
colour and icon selection moved out of it – they remain in the ⋯ menu inside the chat.
## [v202611-00037] – 2026-11-15
### Added
- **Deleting a chat asks about its folder**: if the chat has a folder of its own (inside your folders,
not shared with any other chat, no project folder), you can delete it with the chat – with the
number of files, size and a warning for git repositories. The server checks again before deleting.
## [v202611-00036] – 2026-11-15
### Added
- **Claude Code gets all CodeCraft tools** (image generation, 3D, OpenSCAD, browser, scheduling,
diagnostics, reverse engineering, skills, ask_user, the server's MCP tools …) over an MCP bridge,
as `mcp__codecraft__<name>`. Tools the CLI has itself (read, write, edit, bash, grep …) are not
offered twice. Calls go through the usual approval cards and rules.
## [v202611-00035] – 2026-11-15
### Fixed
- Creating an SSH key failed with "No user exists for uid 99": `ssh-keygen`, `ssh-keyscan` and
`ssh-add` get the same passwd workaround as `ssh`.
## [v202611-00034] – 2026-11-15
### Fixed
- `ssh`, `scp` and `sftp` failed with "No user exists for uid 99" because the container runs as a uid
without passwd entry; the image now wraps them with nss_wrapper (also used by the key wrappers).
## [v202611-00033] – 2026-11-15
### Added
- **SSH keys** (Settings → Account): generate (ed25519) or import keys; agents, the terminal and the
git panel use them automatically for `ssh`, `scp`, `sftp`, `rsync` and git over ssh. Hosts are
trusted on first contact; the public key can be copied for the server's `authorized_keys`.
## [v202611-00032] – 2026-11-15
### Fixed
- Agents now know that git credentials are stored (hosts, no secrets) and no longer search the
environment for them; the system prompt explains `git push` just works and how to read the token
for REST calls without printing it. Claude Code runs get the user's git credentials and identity too.
## [v202611-00031] – 2026-11-15
### Fixed
- Images read by an agent (Read tool) are shown right away; with Claude Code they were dropped.
## [v202611-00030] – 2026-11-15
### Added
- **`/clear`**: the model forgets the conversation so far; the chat keeps showing it, a divider marks
the cut. Also starts a fresh session with Claude Code.
## [v202611-00029] – 2026-11-15
### Added
- **Statistics**: sections and models fold; long history with a year filter (monthly bars for long
ranges); export of the shown figures as XLSX (all tables) or CSV (one table).
- **Prompt snippets**: own text blocks, inserted with `/name` in the message field (Settings → Account).
- **Read-only links**: share a chat's text with people without an account (valid 1–90 days, revocable).
- **Git panel**: the small model suggests a commit message for the staged changes.
- **Weekly review** as a push notification (Mondays 08:00: chats, checks, costs; opt-in).
- **Speech-to-text on the server** (`server.transcribe_model`) for browsers without dictation (Firefox).
- **Offline**: the chat list and the last opened chats stay readable without a connection.
- **Planning model** setting: plan mode always uses it, other modes the chat's model.
- **Automations** send their report as a push notification when a run ends (switchable).
### Fixed
- fal.ai: the provider check no longer reports "API key rejected" when only the model listing is not offered.
## [v202611-00028] – 2026-11-14
### Added
- **18 more chat icons** (gamepad, computer, food, coffee, home, book, code, camera, cart, heart,
star, globe, calendar, car, plane, bulb, money, palette).
### Changed
- **The ⋯ menu of a message sits inside its bubble** (in the foot next to the time, also in your own
bubbles); its popup opens upwards within the bubble and is no wider than the bubble.
## [v202611-00027] – 2026-11-14
### Added
- **Chat bubbles:** your messages on the right in their own colour (the chat's colour tag if it
has one), the agent's answers on the left; tool-only steps and system notes stay without bubble.
- **Icons for chats:** pick one of 22 icons in the ⋯ menu (chat and session list); it shows in the
list, the top bar and the project page, in the chat's colour, and is part of the project default.
### Changed
- **Colour tags are more vivid and more distinct:** eight fixed hues (new: pink), drawn with a
lightness that fits the theme (light/dark) and the highest chroma sRGB allows, so no two tags
look alike; backgrounds are tinted more strongly. Browsers without relative colours keep the
theme's own colours.
## [v202611-00026] – 2026-11-14
### Added
- **Claude Code as an engine:** chats with a "Claude Code" model (Opus, Sonnet, Haiku, default) are
run by the `claude` CLI. Its events become normal chat messages; its tool approvals go through
CodeCraft's permission system (cards, push buttons, "allow for this project", diffs, sandbox).
Sign in from the web UI (Settings → Account → Claude Code: open the sign-in URL, paste the code),
see the CLI version and install updates (admins; installed into the persistent home). The Docker
image now contains the CLI.
- **Small models (8k context) for side tasks:** `small_model` jobs cut their input to the model's
context window. New with it: a one-line status per chat in the session list, "What does this
do?" on approvals, and "Plan as automation" (drafts job instructions from a chat).
- **Mobile:** the top bar no longer pushes the page sideways (extra actions moved into the ⋯ menu);
an image viewer for all images (zoom, swipe between images, mark up and attach to the message,
use in the message); jump to the previous own message; message menu (copy, quote, fork from
here, edit and resend); date and time under every message.
- **Approvals:** push notifications can allow or reject right away; "always allow in this project"
is remembered per folder (never for dangerous commands) and can be revoked in the ⋯ menu.
- **Changes per turn:** files changed by a turn with diffs and "roll back to before this turn".
- **Check after each turn:** a command per project (e.g. `cargo test`) runs after turns that
changed files; green/red badge with output and "let the agent fix it".
- **Colour tags for chats:** tinted chat background and a badge in the list, drawn in the colours
of the active theme.
- **Project defaults, project page, offline queue:** new chats in a folder inherit model, mode,
colour, instructions and folder; a timeline of a project's chats (sidebar → project → ⋯);
messages written without a connection are sent when it is back.
## [v202611-00025] – 2026-11-12
### Fixed
- **Models did not find built-in tools via `mcp_search`** (e.g. `generate_3d` in a chat that
had just been switched from plan to build): the search only looked at MCP servers. It now also
lists matching built-in tools and says whether the current mode has them ("available – call it
directly", or "not in the read-only mode plan – ask the user to switch to build").
## [v202611-00024] – 2026-11-12
### Added
- **3D models with fal.ai (`generate_3d`):** from an image (a project file or one pasted into the
chat, optional back/left/right views), from a text prompt, or from an image steered by a prompt
(the image is refined with the prompt first – the 3D models take no text with an image).
Styles textured, geometry (untextured, for 3D printing) and low-poly, optional PBR. Uses fal.ai's
queue (takes minutes, with progress), saves a GLB plus a preview image, books the cost.
Defaults: Hunyuan3D v3 image-to-3D / text-to-3D (`[images] model_3d`, `model_3d_text`); Tripo
and Trellis endpoints work too.
- The chat card offers "Open in the 3D view"; the media overview lists 3D models (also STL/3MF
written by the agent) with their previews.
## [v202611-00023] – 2026-11-12
### Added
- **Tokens and cost under every answer:** a small line under each answer of the model shows the
tokens sent/received and the cost of all requests of that turn.
### Fixed
- **Total cost of chats stopped growing (opencode Go):** opencode now reports plan requests as
"nothing charged" (cost 0, usage in `allowance`), and CodeCraft took the 0. Plan requests are
valued at the model's list prices again; earlier ones are corrected once on start.
- **Saving opencode Go models failed after fetching prices** ("data did not match any variant of
untagged enum ModelEntry"): opencode lists cache prices per duration for some models
(`{"30m": …, "12h": …}`); the shortest duration is used now.
- The chat title in the top bar no longer shrinks to a single letter when there are many chips.
## [v202611-00022] – 2026-11-11
### Added
- **Theme "Tintenblau"** (dark: ink blue with cream text; light: paper with dark ink), from the
Tintenblau desktop theme; also in the terminal client.
## [v202611-00021] – 2026-11-11
### Added
- **Commits of agents name the model:** every `git commit` an agent makes gets the trailer
`Co-Authored-By: <model> (<provider>) via CodeCraft` (a git wrapper in the agent shell, also in
the sandbox; your own commits stay unchanged). The text is set under *Settings → General*
(`server.commit_trailer`, `{model}`, `{provider}`, `{model_id}`; empty = off).
## [v202611-00020] – 2026-11-11
### Added
- **Reverse engineering outside craft:** *Settings → Agents & commands* can give any mode the
`binary_info` and `decompile` (Ghidra) tools; the clean-room rule stays a craft rule.
- **Raw firmware in `decompile`:** optional `processor` (Ghidra language id, e.g.
`ARM:LE:32:Cortex`) and `base_address` for images without a file format (.bin, converted .hex).
- **Skill "reverse-engineering":** how to identify, decompile and explore programs and firmware
(processor table for common chips, HEX conversion, notes in `analysis/`).
## [v202611-00019] – 2026-11-11
### Added
- **Search in a chat:** the magnifier in the chat's top bar searches all of it (messages, tool
output) on the server; a hit loads its part of the chat and marks the message.
- **Media of a chat:** *⋯ → Media* shows every image, video and audio file of the chat and its
subagents – generated images, screenshots, files the agent wrote, media in answers, your
attachments – filterable, large view, "show in chat" jumps to the message.
### Changed
- **Long chats stay fast:** a chat opens with its messages since the last summary (at most 200);
older ones load when scrolling up, the history before a summary on request. Finished turns
fold into one line ("42 steps · 3 files", the answer stays visible). Typing, streaming and
status changes no longer re-render the whole history, and tool output is only built when a
card is opened. Measured with a chat of 5,900 messages: opening 9 s → 0.7 s (phone-like
CPU: 22 s → 1 s), typing 440 ms → 25 ms per key.
## [v202611-00018] – 2026-11-10
### Added
- **Enforced reviews:** when a review is configured for a mode, every todo the agent marks
completed is reviewed; with findings the todo is reopened and the agent has to fix them and
complete it again (up to the configured rounds per todo). Ending the turn with open todos sends
the agent back to work as long as it makes progress – for unattended runs, e.g. overnight
(such modes may take up to 2,000 steps per turn). The chat shows passed reviews and these
reminders; *Settings → Modes* can switch it off per mode (then only one review at the end of a
turn, as before).
- **Max input per model:** providers that limit the input below the context window (e.g. 272k of
400k) are taken into account; fetched from the model catalogue, editable in the provider settings.
### Fixed
- **Context overflow instead of compaction** (e.g. DeepSeek V4.1 Flash with 384k max output):
requests asked for the model's full max output, and providers count input + output against the
window, so long chats failed before automatic compaction kicked in. The output budget is now what
the window leaves, the chat is compacted when no reasonable room for an answer is left, and a
"context too long" error compacts and retries once instead of stopping. A turn that hits the
step limit now says so.
## [v202611-00017] – 2026-11-10
### Added
- **Markdown in your messages:** your messages are shown formatted like the model's answers
(headings, lists, tables, code blocks, quotes, bold/italic); single line breaks stay as typed.
The eye button next to the text field shows a preview of the formatted message before sending.
## [v202611-00016] – 2026-11-10
### Fixed
- **Phone: page stayed shifted after the keyboard closed** – the top bar (menu) was out of
reach. On Android the keyboard now shrinks the page instead of shifting it
(`interactive-widget=resizes-content`; the chat stays at its latest message); browsers that
still shift the page (iOS Safari) get it scrolled back once the keyboard closes.
## [v202611-00015] – 2026-11-10
### Fixed
- **Generated images missing on other devices:** when an agent moved or renamed a generated
image afterwards (e.g. sorted it into subfolders), the gallery pointed to the old file – only
the device that had it in its browser cache still showed it. Now:
- generated images are also kept in the image store and shown from there when the file is gone
(existing images get their copy on the first start);
- their entries follow moves and renames: after an agent's `mv`/`rename`, when moved with the
file manager, and on the next view (found by file name, or by content when renamed);
- in the chat, an image that cannot be found falls back to the preview the model saw.
## [v202611-00014] – 2026-11-10
### Added
- **New folder in the folder picker:** creating an automation or a chat, *New folder* creates a
subfolder of the shown folder and selects it.
## [v202611-00013] – 2026-11-10
### Added
- **Automation editor:** *New automation* asks for name, project and a starting point (blank
example structure or a template), creates `.codecraft/jobs/<name>.md` and opens an editor:
the instructions in a Markdown editor, schedule and settings next to them, and an assistant
chat in which a model works out the instructions with you and edits the file (reloaded after
each answer). New automations start switched off.
- **Graphical schedule:** every few minutes, hourly, daily, on weekdays, monthly or a cron
expression, with the next runs in server time.
- **Chat list:** folders and projects can be dragged into any order or moved and pinned to the
top via their menu; the order is kept per user on all devices. A new message moves the chat
to the top of its folder or project, no longer the whole project.
### Changed
- **Automation files hold only the instructions;** name, schedule, mode, model, approval mode
and time limit are settings of the automation. Existing job files are taken over on the first
start (settings into the database, the frontmatter removed from the file). Registering a
plain Markdown file creates a switched-off automation to be scheduled in the editor.
## [v202611-00012] – 2026-11-10
### Added
- **Approval mode of automations in the list:** *Automations* shows the approval mode of each
automation as a selection; the change is written into the job file and applies from the next
run. "Allow everything" lets nightly runs work without waiting for approvals (they used to
wait for an answer via Telegram until the timeout).
- **Telegram per chat:** *⋯ → Send questions to Telegram* (shown with a paired bot) decides
whether approvals and questions of that chat and its subagents are sent to Telegram. New chats
from the web UI and automation runs start with it switched off (they stay in the web UI);
chats started from Telegram and existing chats keep reporting. The chat currently open in
Telegram always reports.
## [v202611-00011] – 2026-11-09
### Added
- **Collapsible task list:** the model's task list above the composer folds to one line
(progress and current task) with a click on its heading; the choice is kept per device.
- **Square and short screens** (e.g. Clicks Communicator, phones in landscape): tighter top bar,
composer and new-chat wizard, so the chat keeps most of the height.
### Fixed
- **Blank page on phones after the update:** installed apps (PWA) checked their service worker
at its old URL, got the web page instead of a script and therefore kept the previous app
version forever. With the new image format of v202611-00010 that old version crashed after
showing the chat for a moment. Old service worker URLs now get the clean-up worker, which
removes the stale worker and reloads the current version. Missing scripts/styles answer
404 instead of the web page.
## [v202611-00010] – 2026-11-09
### Changed
- **Chats with many images load much faster:** images (screenshots, generated images, pasted
images) are stored separately instead of inside the messages and load on demand when they
scroll into view, cached by the browser. Large PNGs without transparency are stored as JPEG.
Older chats are converted once on the first start; the database usually shrinks a lot.
Example: a chat with 120 screenshots loads 1.5 MB instead of 66 MB.
- **Compressed responses:** the API and the web UI are sent with gzip/brotli (not images,
downloads or live streams).
## [v202611-00009] – 2026-11-09
### Fixed
- **Sandbox on x86_64:** the start-up check of bubblewrap did not mount `/lib64`, so it failed
on amd64 hosts and shell commands of non-admin users were disabled.
## [v202611-00008] – 2026-11-09
### Added
- **Instructions per chat:** *⋯ → Instructions for this chat* (Telegram: `/prompt`) sets rules
that apply to every message of that chat on top of the mode – language, tone, the part of the
project to work on. A chip in the header shows when a chat has them; forks keep them.
### Fixed
- OpenAI-compatible providers: tool results no longer carry a `name` field. Gateways that
translate to Anthropic (opencode) rejected the request with "anthropic-messages cannot
preserve named conversation participants".
- Telegram: the token field is no longer filled in by the browser's password manager, the bot
token is found in whatever is pasted (also the whole BotFather message), and a wrong format
gets a clear message instead of "HTTP 404".
## [v202611-00007] – 2026-11-09
### Added
- **Telegram bot as a chat interface:** every user can connect their own bot (*Settings →
Account → Telegram*: token from @BotFather, then `/start <code>` pairs it with their Telegram
account – other chats are ignored). Messages, photos and files go to the current session; the
agent's steps show live, the answer comes formatted, images from tools come as photos, and
approvals and questions have buttons. Commands like the terminal client: `/sessions`,
`/session`, `/new`, `/clear`, `/mode`, `/model`, `/permissions`, `/effort`, `/stop`, `/status`,
`/costs`, `/memory`, `/jobs`, `/compact`, `/help`. The server polls Telegram – no public
webhook needed.
## [v202611-00006] – 2026-11-09
### Added
- The image contains the Docker CLI (static client). With the host's `/var/run/docker.sock`
mounted (and its group), admins' agents and terminals can use the host's Docker;
`scripts/move-server.sh --docker-sock` sets this up on a new host.
- `scripts/move-server.sh --tools-dir DIR`: agent home, users' homes and `/tmp` on a separate disk.
## [v202611-00005] – 2026-11-09
### Fixed
- Commits by agents, terminals and the git panel always use the name and e-mail from
*Settings → Account → Git* – also in repositories that have their own `user.name`/`user.email`
(agents had set those as a workaround while no identity was configured). Agents are told not
to change the git identity.
## [v202611-00004] – 2026-11-09
### Added
- **Chats side by side:** *Open beside* in a chat's sidebar menu shows up to three chats as
columns, each with its own tabs and composer (desktop).
- **My modes:** every user can create personal modes under *Settings → My modes* (kind, tools,
read-only, prompt, model); they appear only for that user.
- **Budget warnings** as push notifications for admins (warning share and 100 %) for the total,
provider and image budgets and the opencode Go windows; the status overlay shows all budgets.
- `scripts/move-server.sh` moves an installation to another Docker host with bind mounts
(e.g. Unraid), `scripts/smoke-test.sh` checks a running instance.
### Changed
- File manager tabs of a chat are stored on the server, so they are the same on every device.
## [v202611-00003] – 2026-11-09
### Added
- The Docker image ships the terminal client again (static Linux x86_64 and aarch64 binaries):
`curl -fsSL https://<your-server>/cli/install.sh | sh`, download links in the settings and
`codecraft update` work without building from source.
- File manager: *Open chat here* in a folder's menu starts a chat in that folder.
- **Tabs in a chat:** open terminals and file managers next to the chat (topbar buttons or
"+" in the tab strip), as many as needed. Terminals of a chat keep running on the server until
they are closed, so they are there again after a reload or on another device (output is
replayed); they end when the chat is deleted or the server restarts.
- **Chat mode:** *New session → Just chat* (and mode `chat`) for plain conversation – a lean
prompt with date and personal memory, web search and fetch only, no project or files.
- **Own modes:** *Settings → Agents & commands → New mode* with kind (coding or conversation),
read-only switch, tool checkboxes, prompt, model and review loop; custom modes show up in
*New session* and the mode selector.
- The chat shows when the conversation is being compacted – automatically because the context
is nearly full (with tokens and share of the limit) or on request – and labels the summary.
### Changed
- The mode selector in the composer is a dropdown.
### Removed
- *Settings → Config file* no longer offers to move `${NAME}` values into the configuration (a
one-time migration); `${NAME}` references keep working.
### Fixed
- Session updates from background tasks (title, token usage, phase, todos) and from the UI
no longer overwrite each other. Often the first answer's token usage was lost, so automatic
compaction and the context display started one turn late.
## [v202611-00002] – 2026-11-09
### Added
- **opencode prices:** *Fetch prices* reads prices, context and output limits, image input and
reasoning from opencode's model list (Zen and Go), including long-context tiers and
peak-hour or date-based factors. Costs apply them per request (`price_tiers`,
`price_multipliers` on a model).
- **opencode costs and Go limits:** the cost opencode reports per response is recorded instead
of an estimate, and with a *plan limit* on the provider the Go allowance is tracked in the
5-hour, 7-day and monthly windows (budget table and status overlay; usage through CodeCraft
only).
- **Settings → Config file** lists `${NAME}` references and moves the values of set variables
into the configuration, so `.env` only needs deployment settings.
### Changed
- Examples and forms no longer suggest environment variables for keys; `.env.example` holds
deployment settings only.
## [v202611-00001] – 2026-11-09
### Added
- **Image generation picks the model:** the agent drafts cheap previews, asks which one to
refine (with the price of the final model), then creates the final image with the best model
for the purpose (photo, illustration, text, design). Models that take a reference image
(Nano Banana Pro edit) refine the chosen draft. Known fal.ai models come with sensible
defaults; `stage`, `good_for` and `reference_param` in `[images].models` override them.
- **Restore a backup** in *Settings → Audit*: a stored backup or an uploaded zip. The current
state is backed up first (`…-before-restore.zip`), then CodeCraft restarts with the restored
database and configuration.
- **Test connection** in the provider form checks URL and key before saving.
- **Status LEDs** in the sidebar for server connection, models and MCP servers (green, yellow,
red in vivid variants of the theme colors); a click opens an overview of every provider and
MCP server with its state, latency and tool count.
- `scripts/release.sh`: tag, multi-arch image and Gitea release in one step.
### Changed
- Saving a provider (or reloading providers) checks its status right away instead of at the
next five-minute check.
### Fixed
- opencode Zen/Go: requests carry the session in `x-opencode-session`; opencode Go rejected
them without it ("Request is missing x-opencode-session").
- Two backups in the same second no longer overwrite each other.
## [0.1.0] – 2026-11-09
First public release.
### Agents and chat
- Agent loop with streaming, task lists, subagents (parallel and in the background, on other
models), skills, slash commands, `AGENTS.md` / `CLAUDE.md`, automatic context compaction.
- Modes: **build**, **plan** (read-only) and **craft** (study a website, app, game or old
program and build your own version, with visual comparison, clean room or deep mode),
plus custom modes and subagents. Review loops per mode.
- Tools: files, shell with background processes, web fetch and search, LSP diagnostics after
edits, OpenSCAD for 3D printing, Ghidra for old binaries, image generation (fal.ai), and an
interactive **browser** (click, type, run JavaScript, screenshots, console errors; private
network addresses are blocked).
- MCP servers (stdio and Streamable HTTP, elicitation, tools loaded on demand).
- Permissions per tool and per shell command pattern, modes *ask*, *auto*, *accept edits* and
*bypass*; permission cards with diffs; questions from agents and MCP servers.
- Checkpoints and rewind, fork, regenerate, accept changes hunk by hunk, git worktrees per
session, reasoning effort per message, voice input, `@file` mentions, attachments.
- Project memory (`.codecraft/memory.md`, written by the agent) and personal memory.
- Hooks before and after tool calls, on prompts and at the end of a turn.
### Models
- Anthropic natively and every OpenAI-compatible API (Gemini, OpenRouter, opencode,
fal.ai, Ollama, vLLM, llama.cpp, LM Studio).
- Prices per model, concurrency limits per host and model, monthly budgets.
- **Model fallback** when a host is unreachable or a budget is used up.
- **Provider status**: every 5 minutes each provider's model list is checked (no tokens).
### Web app
- Installable PWA with push notifications (permissions, questions, finished turns) and a
**share target**: text and links shared from other apps start a chat or go into a recent one.
- Guided start screen, file manager with previews (PDF, Office, 3D models, images), web
terminal, image gallery, git panel.
- **Command palette** (Ctrl/Cmd+K) and keyboard shortcuts; **prompt templates** with
placeholders (personal and global); sessions can be **pinned, archived and filed into
folders**, grouped by date or project.
- Statistics: tokens, cache hits, time to first token, prefill and output speed and cost per
request; cost and tokens per day, an activity heatmap of the last 14 days, tables per model,
provider, mode, user and session. Optional **daily cost summary** as a push at 20:00.
- 15 colour themes, each light and dark; English and German.
- "What's new" after updates; admins see when a new release is available.
### Automations and integrations
- **Automations**: agents on a cron schedule from a Markdown job file, created by an agent,
from a **template** or by hand; **incoming webhooks** start them (shared secret or
Gitea/GitHub HMAC signature; the payload is passed to the agent).
- **API tokens** with full or read-only scope.
- **Prometheus metrics** at `/metrics` (own bearer token) and JSON logs
(`CODECRAFT_LOG_FORMAT=json`).
- **Backups**: daily zip of database and config, keep 30, optional copy to any S3-compatible
bucket; download from the settings.
### Terminal client
- `codecraft`: a full-screen terminal client like Claude Code or opencode. The agent loop,
tools and history stay on your machine; models, modes, rules, skills and the theme come
from the server, which counts usage and costs. Also runs **without a server**: a guided
setup connects Anthropic, OpenAI, Gemini, OpenRouter, opencode or local servers (Ollama,
LM Studio, llama.cpp/vLLM, detected automatically).
- No prebuilt binaries in this release: install it from source with
`cargo install --locked --git https://git.familie-zink.org/zink/codecraft codecraft-cli`.
### Users and security
- Local accounts (Argon2) and single sign-on via OpenID Connect; admins and users, private
folders per user, shared folders, read-only session sharing, audit log.
- Shell commands of non-admin users run in a bubblewrap sandbox.
- Multi-arch Docker image (amd64, arm64).
[0.1.0]: https://git.familie-zink.org/zink/codecraft/releases/tag/v0.1.0