Compare commits

..
Author SHA1 Message Date
Richard ZinkandClaude Sonnet 5.5 c82ab5028b README: code factory and skills, fresh screenshots (chat, mobile, settings, statistics, automations, factory, skills)
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-11-17 15:19:54 +01:00
Richard Zink d8c5a0bf0e Release v202611-00058 2026-11-17 15:05:22 +01:00
Richard ZinkandClaude Sonnet 5.5 dad2685ea3 Skills labels without plural problems
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-11-17 15:05:21 +01:00
Richard Zink 84ee68d4da Release v202611-00057 2026-11-17 14:43:39 +01:00
Richard ZinkandClaude Sonnet 5.5 43b07aca48 Factory: skills per phase
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
2026-11-17 14:43:38 +01:00
14 changed files with 103 additions and 13 deletions

No files matched your search

+13
View File
@@ -3,6 +3,19 @@
All notable changes to CodeCraft. Versions follow [Semantic Versioning](https://semver.org/); All notable changes to CodeCraft. Versions follow [Semantic Versioning](https://semver.org/);
until 1.0, minor versions may contain breaking changes. until 1.0, minor versions may contain breaking changes.
## [v202611-00058] – 2026-11-17
### Changed
- **README and screenshots** brought up to date (code factory, skills, new interface); skills labels no longer say "1 files".
## [v202611-00057] – 2026-11-17
### Added
- **Factory: skills per phase.** In the repository settings, each phase (review, build, acceptance) can have
skills switched on; the chat of that phase follows them in every turn.
## [v202611-00056] – 2026-11-17 ## [v202611-00056] – 2026-11-17
### Added ### Added
+25 -6
View File
@@ -11,7 +11,7 @@
</p> </p>
<p align="center"> <p align="center">
<img alt="Version v202611-00056" src="https://img.shields.io/badge/version-v202611--00056-d69e2e?style=flat-square"> <img alt="Version v202611-00058" src="https://img.shields.io/badge/version-v202611--00058-d69e2e?style=flat-square">
<img alt="License: MIT" src="https://img.shields.io/badge/license-MIT-689d6a?style=flat-square"> <img alt="License: MIT" src="https://img.shields.io/badge/license-MIT-689d6a?style=flat-square">
<img alt="Rust backend" src="https://img.shields.io/badge/backend-Rust-b7410e?style=flat-square&logo=rust"> <img alt="Rust backend" src="https://img.shields.io/badge/backend-Rust-b7410e?style=flat-square&logo=rust">
<img alt="React PWA" src="https://img.shields.io/badge/frontend-React%20PWA-2d81ff?style=flat-square&logo=react"> <img alt="React PWA" src="https://img.shields.io/badge/frontend-React%20PWA-2d81ff?style=flat-square&logo=react">
@@ -37,7 +37,7 @@
<br> <br>
<p align="center"> <p align="center">
<img src="docs/images/chat.png" alt="A CodeCraft session in the Gruvbox dark theme: a diff of the fix, the task list, and a permission card asking before the test suite runs; the sidebar shows pinned chats, a folder and the archive" width="100%"> <img src="docs/images/chat.png" alt="A CodeCraft session in the Gruvbox dark theme: the agent's task list and a permission card with the diff of a one-line fix; the sidebar shows a pinned chat, a folder and today's chats, and the chip in the top bar shows an active skill" width="100%">
<br> <br>
<sub>A session: the agent works through its task list, shows diffs, and asks before anything outside the rules.</sub> <sub>A session: the agent works through its task list, shows diffs, and asks before anything outside the rules.</sub>
</p> </p>
@@ -130,7 +130,7 @@ opens the command palette, and prompt templates with placeholders save typing.
<table> <table>
<tr> <tr>
<td width="40%" valign="top"> <td width="40%" valign="top">
<img src="docs/images/mobile.png" alt="CodeCraft on a phone in a light theme: the agent asks where coupon codes should be stored and offers three options plus free text" width="100%"><br> <img src="docs/images/mobile.png" alt="CodeCraft on a phone in a dark theme: the task list and a permission card with the diff, with Allow once, Always allow and Reject buttons" width="100%"><br>
<sub><b>Mobile:</b> answer questions and approve changes from your phone. Share links and text from other apps into a chat.</sub> <sub><b>Mobile:</b> answer questions and approve changes from your phone. Share links and text from other apps into a chat.</sub>
</td> </td>
<td width="60%" valign="top"> <td width="60%" valign="top">
@@ -154,6 +154,19 @@ slash commands in `.claude/commands` and skills in `.claude/skills` are picked u
The agent keeps notes about a project in `.codecraft/memory.md`; your personal memory applies to The agent keeps notes about a project in `.codecraft/memory.md`; your personal memory applies to
every project. every project.
<p align="center">
<img src="docs/images/skills.png" alt="Settings, Skills tab: the user's own skills with description and file count, buttons for a new skill and the import from GitHub" width="100%">
</p>
**Skills** are packaged instructions for a kind of task: a folder with a `SKILL.md` (name, description,
instructions) and further files. Create them in *Settings → Skills* with a small editor, or import them
from a GitHub repository or folder (every folder with a `SKILL.md` is offered; filter and select several).
They are stored per user and work next to the skills of your projects and the built-in ones. To keep the
context small, you switch skills on **per chat** (menu ⋯ → *Skills for this chat*): only those are part of
every turn. All others cost nothing – up to 12 are listed by name, a larger library is not listed at all and
the agent finds a skill by keyword and loads it, with its further files, when it needs it. In the
[code factory](#code-factory) each phase can have its own skills.
## Models ## Models
Every provider is an endpoint with a list of models. Each model can have a context size, Every provider is an endpoint with a list of models. Each model can have a context size,
@@ -204,7 +217,7 @@ platforms: `cargo install --locked --git https://git.familie-zink.org/zink/codec
## Automations ## Automations
<p align="center"> <p align="center">
<img src="docs/images/automations.png" alt="Creating an automation from a template: schedule chips, cron expression, mode, permission mode, model and the task for the agent" width="100%"> <img src="docs/images/automations.png" alt="The automation editor: schedule chips with the next runs, mode, approval mode, model, time limit and the instructions file next to an assistant chat" width="100%">
</p> </p>
An automation is an agent that starts on a schedule. Its instructions are a Markdown file An automation is an agent that starts on a schedule. Its instructions are a Markdown file
@@ -226,6 +239,12 @@ The factory turns issues on your Git forge (Gitea, GitHub or GitLab) into review
merged changes – with agents doing the work and you deciding at the points that matter. It is merged changes – with agents doing the work and you deciding at the points that matter. It is
the **Factory** page in the sidebar; all chats it creates live in the sidebar folder *Factory*. the **Factory** page in the sidebar; all chats it creates live in the sidebar folder *Factory*.
<p align="center">
<img src="docs/images/factory.png" alt="The factory board: columns for review, questions, approved, build, acceptance, failed and done with issue cards, the repository tile with its version, and buttons to plan a feature, cut a milestone and start a release" width="100%">
<br>
<sub>The factory page: every issue sits in the column of its phase; the repository shows its current version.</sub>
</p>
``` ```
issue + label "factory" issue + label "factory"
│ │
@@ -291,8 +310,8 @@ the files attached (Gitea and GitHub).
**Settings per repository** (*Factory → Repositories*, *New repository*): forge connection and token, label **Settings per repository** (*Factory → Repositories*, *New repository*): forge connection and token, label
(default `factory`), parallel runs, allowed issue authors (default: the token's owner), and per (default `factory`), parallel runs, allowed issue authors (default: the token's owner), and per
phase the model, agent (CodeCraft or Claude Code), permission mode, timeout and tries on phase the model, agent (CodeCraft or Claude Code), permission mode, timeout, tries on
technical failures, plus the number of rework rounds (default 2). Failed or rejected issues technical failures, the skills that are switched on for it, plus the number of rework rounds (default 2). Failed or rejected issues
can be restarted from the board, finished ones are closed with a comment. can be restarted from the board, finished ones are closed with a comment.
**Files in the repository** (all optional, kept in `.factory/`): **Files in the repository** (all optional, kept in `.factory/`):
Binary file not shown.

Before

Width:  |  Height:  |  Size: 118 KiB

After

Width:  |  Height:  |  Size: 167 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 144 KiB

After

Width:  |  Height:  |  Size: 156 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 136 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 172 KiB

After

Width:  |  Height:  |  Size: 176 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 102 KiB

After

Width:  |  Height:  |  Size: 123 KiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 134 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 177 KiB

After

Width:  |  Height:  |  Size: 203 KiB

+7 -2
View File
@@ -49,6 +49,9 @@ pub struct PhaseCfg {
/// Extra instructions appended to the phase prompt. /// Extra instructions appended to the phase prompt.
#[serde(default)] #[serde(default)]
pub extra_prompt: String, pub extra_prompt: String,
/// Skills (names) switched on for the chat of this phase; only the CodeCraft agent uses them.
#[serde(default)]
pub skills: Vec<String>,
} }
fn default_agent() -> String { fn default_agent() -> String {
@@ -69,7 +72,7 @@ fn default_rework() -> i64 {
impl Default for PhaseCfg { impl Default for PhaseCfg {
fn default() -> Self { fn default() -> Self {
Self { model: String::new(), agent: default_agent(), permission_mode: default_mode(), timeout_minutes: default_timeout(), max_attempts: default_attempts(), extra_prompt: String::new() } Self { model: String::new(), agent: default_agent(), permission_mode: default_mode(), timeout_minutes: default_timeout(), max_attempts: default_attempts(), extra_prompt: String::new(), skills: Vec::new() }
} }
} }
@@ -171,6 +174,7 @@ impl RepoCfg {
} }
p.timeout_minutes = p.timeout_minutes.clamp(1, 24 * 60); p.timeout_minutes = p.timeout_minutes.clamp(1, 24 * 60);
p.max_attempts = p.max_attempts.clamp(1, 10); p.max_attempts = p.max_attempts.clamp(1, 10);
p.skills = p.skills.iter().map(|n| n.trim().to_string()).filter(|n| !n.is_empty() && n.len() <= 64).take(10).collect();
} }
Ok(()) Ok(())
} }
@@ -881,7 +885,8 @@ fn start_phase(app: &Arc<App>, user: &User, cfg: &RepoCfg, issue: i64, phase: &s
telegram_quiet: true, telegram_quiet: true,
summary: String::new(), summary: String::new(),
color: None, color: None,
icon: None, skills: Vec::new(), icon: None,
skills: pc.skills.clone(),
}; };
let session = app.save_session(session)?; let session = app.save_session(session)?;
let run_id = crate::types::new_id(); let run_id = crate::types::new_id();
+38 -1
View File
@@ -30,6 +30,7 @@ interface PhaseCfg {
timeout_minutes: number; timeout_minutes: number;
max_attempts: number; max_attempts: number;
extra_prompt: string; extra_prompt: string;
skills: string[];
} }
interface RepoCfg { interface RepoCfg {
url: string; url: string;
@@ -575,11 +576,46 @@ function FactoryOverview({ onSettings }: { onSettings: () => void }) {
); );
} }
const blankPhase = (): PhaseCfg => ({ model: "", agent: "codecraft", permission_mode: "auto", timeout_minutes: 60, max_attempts: 2, extra_prompt: "" }); const blankPhase = (): PhaseCfg => ({ model: "", agent: "codecraft", permission_mode: "auto", timeout_minutes: 60, max_attempts: 2, extra_prompt: "", skills: [] });
const blankRepo = (connection = "", kind = "gitea"): RepoCfg => ({ url: "", kind, connection, token: "", label: "factory", max_parallel: 1, max_rework: 2, allowed_users: [], enabled: true, review: blankPhase(), build: blankPhase(), qa: blankPhase() }); const blankRepo = (connection = "", kind = "gitea"): RepoCfg => ({ url: "", kind, connection, token: "", label: "factory", max_parallel: 1, max_rework: 2, allowed_users: [], enabled: true, review: blankPhase(), build: blankPhase(), qa: blankPhase() });
const blankConn = (): Connection => ({ id: "", name: "", kind: "gitea", url: "", token: "", owner: "" }); const blankConn = (): Connection => ({ id: "", name: "", kind: "gitea", url: "", token: "", owner: "" });
const MODES = ["default", "auto", "accept_edits", "bypass"]; const MODES = ["default", "auto", "accept_edits", "bypass"];
/** Skills switched on for the chat of one phase (the agent follows them in every turn). */
function PhaseSkills({ value, onChange }: { value: string[]; onChange: (v: string[]) => void }) {
const t = useT();
const [all, setAll] = useState<{ name: string; description: string }[]>([]);
useEffect(() => {
get<{ skills: { name: string; description: string }[] }>("/commands")
.then((d) => setAll(d.skills))
.catch(() => {});
}, []);
const rest = all.filter((s) => !value.includes(s.name));
return (
<div className="factory-skills">
<span>{t("factory.skills")}</span>
<div className="row" style={{ gap: 4, flexWrap: "wrap" }}>
{value.map((n) => (
<button key={n} type="button" className="chip" style={{ border: "none" }} title={t("factory.skillRemove")} onClick={() => onChange(value.filter((x) => x !== n))}>
{n} ×
</button>
))}
{rest.length > 0 && value.length < 10 && (
<select value="" onChange={(e) => e.target.value && onChange([...value, e.target.value])}>
<option value="">{t("factory.skillAdd")}</option>
{rest.map((s) => (
<option key={s.name} value={s.name} title={s.description}>
{s.name}
</option>
))}
</select>
)}
</div>
<div className="hint">{t("factory.skillsHint")}</div>
</div>
);
}
function PhaseForm({ phase, value, onChange }: { phase: Phase; value: PhaseCfg; onChange: (patch: Partial<PhaseCfg>) => void }) { function PhaseForm({ phase, value, onChange }: { phase: Phase; value: PhaseCfg; onChange: (patch: Partial<PhaseCfg>) => void }) {
const t = useT(); const t = useT();
return ( return (
@@ -614,6 +650,7 @@ function PhaseForm({ phase, value, onChange }: { phase: Phase; value: PhaseCfg;
<span>{t("factory.attempts")}</span> <span>{t("factory.attempts")}</span>
<input type="number" min={1} max={10} value={value.max_attempts} title={t("factory.attemptsHint")} onChange={(e) => onChange({ max_attempts: Math.max(1, Number(e.target.value) || 1) })} /> <input type="number" min={1} max={10} value={value.max_attempts} title={t("factory.attemptsHint")} onChange={(e) => onChange({ max_attempts: Math.max(1, Number(e.target.value) || 1) })} />
</label> </label>
<PhaseSkills value={value.skills ?? []} onChange={(skills) => onChange({ skills })} />
<label> <label>
<span>{t("factory.extraPrompt")}</span> <span>{t("factory.extraPrompt")}</span>
<textarea rows={3} value={value.extra_prompt} onChange={(e) => onChange({ extra_prompt: e.target.value })} /> <textarea rows={3} value={value.extra_prompt} onChange={(e) => onChange({ extra_prompt: e.target.value })} />
+6 -2
View File
@@ -790,7 +790,7 @@ export const de: Messages = {
"skills.skipped": "übersprungen", "skills.skipped": "übersprungen",
"skills.filter": "Filtern…", "skills.filter": "Filtern…",
"skills.empty": "Noch keine Skills.", "skills.empty": "Noch keine Skills.",
"skills.files": "{n} Dateien", "skills.files": "Dateien: {n}",
"skills.name": "Name (Ordner, Kleinbuchstaben)", "skills.name": "Name (Ordner, Kleinbuchstaben)",
"skills.file": "Datei", "skills.file": "Datei",
"skills.addFile": "Pfad der neuen Datei", "skills.addFile": "Pfad der neuen Datei",
@@ -799,7 +799,7 @@ export const de: Messages = {
"skills.chatTitle": "Skills für diesen Chat", "skills.chatTitle": "Skills für diesen Chat",
"skills.chatHint": "Eingeschaltete Skills gelten für jede Nachricht dieses Chats. Schalte nur ein, was du brauchst – jeder belegt Kontext.", "skills.chatHint": "Eingeschaltete Skills gelten für jede Nachricht dieses Chats. Schalte nur ein, was du brauchst – jeder belegt Kontext.",
"skills.manyActive": "Viele aktive Skills belegen viel Kontext.", "skills.manyActive": "Viele aktive Skills belegen viel Kontext.",
"skills.chip": "{n} Skills", "skills.chip": "Skills: {n}",
"instructions.chip": "Anweisungen", "instructions.chip": "Anweisungen",
"chat.compacted": "Unterhaltung komprimiert – Zusammenfassung", "chat.compacted": "Unterhaltung komprimiert – Zusammenfassung",
"chat.context": "Kontext", "chat.context": "Kontext",
@@ -1292,6 +1292,10 @@ export const de: Messages = {
"factory.closeMsg": "Kommentar zum Schließen von Issue #{n}. Er wird im Issue veröffentlicht, danach wird das Issue geschlossen.", "factory.closeMsg": "Kommentar zum Schließen von Issue #{n}. Er wird im Issue veröffentlicht, danach wird das Issue geschlossen.",
"factory.closeDefault": "Erledigt und abgenommen.", "factory.closeDefault": "Erledigt und abgenommen.",
"factory.closed": "Issue geschlossen", "factory.closed": "Issue geschlossen",
"factory.skills": "Skills",
"factory.skillAdd": "Skill hinzufügen…",
"factory.skillRemove": "Entfernen",
"factory.skillsHint": "Für jeden Chat dieser Phase eingeschaltet (CodeCraft-Agent). Jeder belegt Kontext – wenige wählen.",
"factory.pr": "Pull Request", "factory.pr": "Pull Request",
"factory.round": "Runde {n}", "factory.round": "Runde {n}",
"factory.openChat": "Chat öffnen", "factory.openChat": "Chat öffnen",
+6 -2
View File
@@ -789,7 +789,7 @@ export const en = {
"skills.skipped": "skipped", "skills.skipped": "skipped",
"skills.filter": "Filter…", "skills.filter": "Filter…",
"skills.empty": "No skills yet.", "skills.empty": "No skills yet.",
"skills.files": "{n} files", "skills.files": "Files: {n}",
"skills.name": "Name (folder, lowercase)", "skills.name": "Name (folder, lowercase)",
"skills.file": "File", "skills.file": "File",
"skills.addFile": "Path of the new file", "skills.addFile": "Path of the new file",
@@ -798,7 +798,7 @@ export const en = {
"skills.chatTitle": "Skills for this chat", "skills.chatTitle": "Skills for this chat",
"skills.chatHint": "Switched-on skills are part of every message of this chat. Switch on only what you need – each one uses context.", "skills.chatHint": "Switched-on skills are part of every message of this chat. Switch on only what you need – each one uses context.",
"skills.manyActive": "Many active skills use a lot of context.", "skills.manyActive": "Many active skills use a lot of context.",
"skills.chip": "{n} skills", "skills.chip": "Skills: {n}",
"instructions.chip": "Instructions", "instructions.chip": "Instructions",
"chat.compacted": "Conversation compacted – summary", "chat.compacted": "Conversation compacted – summary",
"chat.context": "Context", "chat.context": "Context",
@@ -1291,6 +1291,10 @@ export const en = {
"factory.closeMsg": "Comment for closing issue #{n}. It is posted on the issue, then the issue is closed.", "factory.closeMsg": "Comment for closing issue #{n}. It is posted on the issue, then the issue is closed.",
"factory.closeDefault": "Done and accepted.", "factory.closeDefault": "Done and accepted.",
"factory.closed": "Issue closed", "factory.closed": "Issue closed",
"factory.skills": "Skills",
"factory.skillAdd": "Add skill…",
"factory.skillRemove": "Remove",
"factory.skillsHint": "Switched on for every chat of this phase (CodeCraft agent). Each one uses context – choose few.",
"factory.pr": "Pull request", "factory.pr": "Pull request",
"factory.round": "round {n}", "factory.round": "round {n}",
"factory.openChat": "Open chat", "factory.openChat": "Open chat",
+8
View File
@@ -3,6 +3,14 @@
* under "unreleased"; scripts/release.sh renames the key to the release version. * under "unreleased"; scripts/release.sh renames the key to the release version.
*/ */
export const WHATS_NEW: Record<string, { en: string[]; de: string[] }> = { export const WHATS_NEW: Record<string, { en: string[]; de: string[] }> = {
"202611-00058": {
en: ["README and screenshots updated"],
de: ["README und Screenshots aktualisiert"],
},
"202611-00057": {
en: ["Factory: choose skills for each phase (review, build, acceptance)"],
de: ["Factory: Skills für jede Phase wählbar (Review, Build, Abnahme)"],
},
"202611-00056": { "202611-00056": {
en: ["Skills: create or import them from GitHub (Settings → Skills) and switch them on per chat; large libraries do not fill the context"], en: ["Skills: create or import them from GitHub (Settings → Skills) and switch them on per chat; large libraries do not fill the context"],
de: ["Skills: selbst anlegen oder von GitHub importieren (Einstellungen → Skills) und pro Chat einschalten; große Bibliotheken füllen den Kontext nicht"], de: ["Skills: selbst anlegen oder von GitHub importieren (Einstellungen → Skills) und pro Chat einschalten; große Bibliotheken füllen den Kontext nicht"],