sync: upgrade upstream GSD from v1.22.4 to v1.30.0

CHANGELOG.md

@@ -5,6 +5,58 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [1.22.2] - 2026-03-30
+
+Overview: Major upstream sync from GSD v1.30.0 adding autonomous execution, fast mode, UI design pipeline, multi-project workspaces, user profiling, forensics, and 25 new slash commands. Full documentation now available in four additional locales (ja-JP, ko-KR, pt-BR, zh-CN). Added `mode: subagent` declarations to all agent definition files.
+
+### Added
+
+- `/gsd-autonomous` command and `autonomous` workflow for end-to-end autonomous project execution
+- `/gsd-fast` command and `fast` workflow for rapid-fire task execution
+- `/gsd-do` command and `do` workflow for direct task execution without full planning
+- `/gsd-forensics` command and `forensics` workflow for post-mortem investigation and debugging
+- `/gsd-thread` command for threaded conversation management
+- `/gsd-workstreams` command and `workstream.cjs` library for multi-stream work management with `--workstream` flag
+- `/gsd-profile-user` command, `profile-user` workflow, `gsd-user-profiler` agent, `profile-pipeline.cjs`, and `profile-output.cjs` for building and maintaining user preference profiles
+- `/gsd-ui-phase` command, `ui-phase` workflow, `gsd-ui-researcher` agent, `gsd-ui-auditor` agent, `gsd-ui-checker` agent for complete UI design and review pipeline
+- `/gsd-ui-review` command and `ui-review` workflow for retroactive UI audits
+- `/gsd-ship` command and `ship` workflow for streamlined shipping and release
+- `/gsd-pr-branch` command and `pr-branch` workflow for PR branch creation and management
+- `/gsd-manager` command and `manager` workflow for manager-level project oversight
+- `/gsd-session-report` command and `session-report` workflow for session summary generation
+- `/gsd-stats` command and `stats` workflow for project and milestone statistics
+- `/gsd-note` command and `note` workflow for quick note-taking during work
+- `/gsd-add-backlog` and `/gsd-review-backlog` commands for backlog item management
+- `/gsd-milestone-summary` command and `milestone-summary` workflow for milestone overview generation
+- `/gsd-new-workspace`, `/gsd-list-workspaces`, and `/gsd-remove-workspace` commands and workflows for multi-project workspace management
+- `/gsd-next` command and `next` workflow for determining and starting the next task
+- `/gsd-plant-seed` command and `plant-seed` workflow for seeding new project ideas
+- `/gsd-audit-uat` command, `audit-uat` workflow, and `uat.cjs` library for user acceptance testing audits
+- `/gsd-review` command and `review` workflow for structured code review
+- `gsd-advisor-researcher` agent for researching gray area decisions with comparison tables
+- `gsd-assumptions-analyzer` agent and `discuss-phase-assumptions` workflow for assumption analysis during discuss phase
+- `model-profiles.cjs` library for model profile definitions
+- `security.cjs` library for security utility functions
+- `node-repair` workflow for automated Node.js dependency repair
+- `UI-SPEC.md`, `claude-md.md`, `dev-preferences.md`, `discussion-log.md`, and `user-profile.md` templates
+- `user-profiling.md` and `workstream-flag.md` reference documentation
+- `superpowers/plans/2026-03-18-materialize-new-project-config.md` and `superpowers/specs/2026-03-20-multi-project-workspaces-design.md` design documents
+- Skill definitions (SKILL.md) for gsd-audit-milestone, gsd-cleanup, gsd-complete-milestone, gsd-discuss-phase, gsd-execute-phase, gsd-plan-phase, gsd-ui-phase, gsd-ui-review, and gsd-verify-work
+- `mode: subagent` declarations added to all 18 agent definition files in `gsd-opencode/agents/`
+- Full Japanese (ja-JP) documentation: AGENTS, ARCHITECTURE, CLI-TOOLS, COMMANDS, CONFIGURATION, FEATURES, README, USER-GUIDE, context-monitor, and workflow-discuss-mode
+- Full Korean (ko-KR) documentation: AGENTS, ARCHITECTURE, CLI-TOOLS, COMMANDS, CONFIGURATION, FEATURES, README, USER-GUIDE, context-monitor, and workflow-discuss-mode
+- Brazilian Portuguese (pt-BR) documentation: AGENTS, ARCHITECTURE, CLI-TOOLS, COMMANDS, CONFIGURATION, FEATURES, README, USER-GUIDE, context-monitor, and workflow-discuss-mode
+- Simplified Chinese (zh-CN) documentation: README, USER-GUIDE, and full references subdirectory (checkpoints, continuation-format, decimal-phase-calculation, git-integration, git-planning-commit, model-profile-resolution, model-profiles, phase-argument-parsing, planning-config, questioning, tdd, ui-brand, verification-patterns)
+
+### Changed
+
+- Updated 12 existing agents (codebase-mapper, debugger, executor, integration-checker, nyquist-auditor, phase-researcher, plan-checker, planner, project-researcher, research-synthesizer, roadmapper, verifier) with improved behavior
+- Enhanced 8 existing commands (discuss-phase, execute-phase, plan-phase, research-phase, debug, quick, reapply-patches, set-profile)
+- Improved 30 existing workflows including execute-phase, plan-phase, discuss-phase, new-project, new-milestone, health, help, settings, and update
+- Updated 12 CLI library modules (commands, config, core, frontmatter, init, milestone, phase, roadmap, state, template, verify, gsd-tools) for v1.30.0 compatibility
+- Modified templates: config.json, context.md, phase-prompt.md, project.md, and UAT.md
+- Updated 7 reference documents: checkpoints, decimal-phase-calculation, git-integration, model-profile-resolution, model-profiles, phase-argument-parsing, and planning-config
+
 ## [1.22.0] - 2026-03-08
 
 Overview: Synchronized with upstream GSD v1.22.4 to fix agent execution syntax and prevent unexpected stops. Simplified model profile system from quality/balanced/budget to simple/smart/genius with updated configuration file structure. Enhanced copy and translate services for upstream synchronization.

README.md

@@ -1,6 +1,6 @@
 <div align="left">
 
-# GET SHIT DONE for OpenCode. (Based on TÂCHES v1.22.4 - 2026-03-03)
+# GET SHIT DONE for OpenCode. (Based on TÂCHES v1.30.0 - 2026-03-30)
 
 **A light-weight and powerful meta-prompting, context engineering and spec-driven development system for Claude Code by TÂCHES. (Adapted for OpenCode by rokicool and enthusiasts)**
 
@@ -87,6 +87,12 @@ I just love both GSD and OpenCode. I felt like having GSD available only for Cla
 
 — **Roman**
 
+## Version 1.30.0
+
+We are keeping up with original GSDv1 v1.30.0 (2026-03-30)
+
+There is a controvertial solution I used to support "skill(skill='command-name')" syntax. Apart from that everything is expected to be fully functional.
+
 ## Version 1.22.1
 
 I decided to add 'mode: subagent' property to all custom agents. It should not affect any GSD functionality. However, it should remove unnecessary agents out of the list, available by Tab.

assets/bin/gsd-translate-in-place.js

@@ -19,8 +19,8 @@
  *   2  Runtime error (file I/O, permissions)
  */
 
-import { readFile, writeFile, access } from 'node:fs/promises';
-import { resolve, dirname } from 'node:path';
+import { readFile, writeFile, access, mkdir } from 'node:fs/promises';
+import { resolve, dirname, basename } from 'node:path';
 import { fileURLToPath } from 'node:url';
 
 // Use dynamic import for tinyglobby
@@ -269,6 +269,159 @@ async function loadConfig(configPath) {
   return loadConfigs([configPath]);
 }
 
+async function ensureCommandNames(apply) {
+  const commandsDir = resolve(__dirname, '../../gsd-opencode/commands/gsd');
+
+  let commandFiles;
+  try {
+    commandFiles = await glob(['*.md'], { cwd: commandsDir, onlyFiles: true, absolute: true });
+  } catch {
+    console.log('No command files found to check for missing name: field.');
+    return { fixed: 0, missing: 0 };
+  }
+
+  if (!commandFiles || commandFiles.length === 0) {
+    console.log('No command files found to check for missing name: field.');
+    return { fixed: 0, missing: 0 };
+  }
+
+  let fixed = 0;
+  let missing = 0;
+
+  for (const filePath of commandFiles) {
+    const commandName = basename(filePath, '.md');
+    let content;
+
+    try {
+      content = await readFile(filePath, 'utf-8');
+    } catch {
+      continue;
+    }
+
+    const frontmatterMatch = content.match(/^---\n([\s\S]*?)\n---/);
+    if (!frontmatterMatch) continue;
+
+    const frontmatter = frontmatterMatch[1];
+
+    if (/^name:/m.test(frontmatter)) continue;
+
+    missing++;
+    const newFrontmatter = `name: ${commandName}\n${frontmatter}`;
+    const newContent = content.replace(
+      /^---\n([\s\S]*?)\n---/,
+      `---\n${newFrontmatter}\n---`
+    );
+
+    if (apply) {
+      await writeFile(filePath, newContent, 'utf-8');
+      console.log(`  Fixed missing name: in ${commandName}.md`);
+      fixed++;
+    } else {
+      console.log(`  [dry-run] Would add name: ${commandName} to ${commandName}.md`);
+    }
+  }
+
+  if (missing > 0) {
+    console.log(`Command name check: ${missing} file(s) missing name: field${apply ? `, ${fixed} fixed` : ' (dry-run)'}`);
+  } else {
+    console.log('All command files have name: in frontmatter.');
+  }
+
+  return { fixed, missing };
+}
+
+async function discoverGsdSkillReferences(searchPatterns) {
+  const files = await glob(searchPatterns, {
+    ignore: ['node_modules/**', '.git/**'],
+    onlyFiles: true
+  });
+
+  if (!files || files.length === 0) {
+    return new Set();
+  }
+
+  const skillRefs = new Set();
+  const pattern = /skill\(skill="gsd-([^"]+)"/g;
+
+  for (const file of files) {
+    try {
+      const content = await readFile(file, 'utf-8');
+      let match;
+      while ((match = pattern.exec(content)) !== null) {
+        skillRefs.add(`gsd-${match[1]}`);
+      }
+      pattern.lastIndex = 0;
+    } catch {
+      // skip unreadable files
+    }
+  }
+
+  return skillRefs;
+}
+
+async function generateSkillWrappers(commandNames) {
+  const commandsDir = resolve(__dirname, '../../gsd-opencode/commands/gsd');
+  const skillsBaseDir = resolve(__dirname, '../../gsd-opencode/skills');
+
+  const commandSet = new Set(commandNames);
+  let created = 0;
+  let skipped = 0;
+  let notFound = 0;
+
+  for (const commandName of commandSet) {
+    const commandFile = resolve(commandsDir, `${commandName}.md`);
+    const skillDir = resolve(skillsBaseDir, commandName);
+    const skillFile = resolve(skillDir, 'SKILL.md');
+
+    try {
+      await access(commandFile);
+    } catch {
+      console.warn(`  Command file not found: ${commandName}.md`);
+      notFound++;
+      continue;
+    }
+
+    try {
+      await access(skillFile);
+      skipped++;
+      continue;
+    } catch {
+      // file doesn't exist yet, proceed
+    }
+
+    const content = await readFile(commandFile, 'utf-8');
+
+    const frontmatterMatch = content.match(/^---\n([\s\S]*?)\n---\n?([\s\S]*)$/);
+    let skillContent;
+
+    if (frontmatterMatch) {
+      const originalFm = frontmatterMatch[1].trim();
+      const body = frontmatterMatch[2];
+
+      const nameMatch = originalFm.match(/^name:\s*(.+)$/m);
+      const commandDisplayName = nameMatch ? nameMatch[1].trim() : commandName;
+
+      const newFrontmatter = [
+        `---`,
+        `name: ${commandName}`,
+        `description: Implementation of ${commandDisplayName} command`,
+        `---`
+      ].join('\n');
+
+      skillContent = newFrontmatter + '\n\n' + body;
+    } else {
+      skillContent = `---\nname: ${commandName}\ndescription: Implementation of ${commandName} command\n---\n\n` + content;
+    }
+
+    await mkdir(skillDir, { recursive: true });
+    await writeFile(skillFile, skillContent, 'utf-8');
+    created++;
+  }
+
+  console.log(`Skill wrappers: created ${created}, skipped ${skipped} (already exist), not found ${notFound}`);
+  return { created, skipped, notFound };
+}
+
 /**
  * Main execution
  */
@@ -483,6 +636,19 @@ async function main() {
   }
 
   console.log('');
+
+  console.log('Checking command files for missing name: in frontmatter...');
+  await ensureCommandNames(args.apply);
+  console.log('');
+
+  const skillRefs = await discoverGsdSkillReferences(config.patterns);
+  if (skillRefs.size > 0) {
+    console.log(`Discovered ${skillRefs.size} gsd skill reference(s): ${[...skillRefs].join(', ')}`);
+    await generateSkillWrappers([...skillRefs]);
+  } else {
+    console.log('No gsd skill references found in processed files.');
+  }
+
   console.log(formatter.formatSuccess('Done!'));
   process.exit(EXIT_SUCCESS);
 }

assets/configs/config.json

@@ -264,13 +264,21 @@
       "description": "Generic regex to convert any remaining tools list to YAML map"
     },
     {
-      "_comment": "Generic regex fallback for any remaining tools: lines",
+      "_comment": "allowed-tools with inline tool list (e.g. 'allowed-tools: Read, Write, Edit')",
+      "pattern": "^allowed-tools: ([A-Za-z, ]+)$",
+      "replacement": "permissions",
+      "caseSensitive": true,
+      "isRegex": true,
+      "transform": "allowed_tools_inline_to_yaml",
+      "description": "Convert inline allowed-tools list to YAML permissions map"
+    },
+    {
+      "_comment": "allowed-tools without inline list (multiline, already handled by - Read rules)",
       "pattern": "allowed-tools:",
       "replacement": "permissions:",
       "caseSensitive": true,
       "isRegex": false,
-      "transform": "allowed-tools to permissions",
-      "description": ""
+      "description": "Rename allowed-tools key to permissions for multiline format"
     },
     {
       "pattern": "<sub>",
@@ -320,6 +328,12 @@
       "caseSensitive": true,
       "description": "Transform named color purple to hex"
     },
+    {
+      "pattern": "color: magenta",
+      "replacement": "color: \"#FF00FF\"",
+      "caseSensitive": true,
+      "description": "Transform named color purple to hex"
+    },
     {
       "pattern": "All arguments become",
       "replacement": "$ARGUMENTS become",
@@ -332,6 +346,16 @@
       "caseSensitive": true,
       "description": "Transform general-purpose subagent type to general"
     },
+    {
+      "pattern": "'general-purpose'",
+      "replacement": "'general'",
+      "description": "Fix: quoted general-purpose subagent type name to general"
+    },
+    {
+      "pattern": "general-purpose research agent",
+      "replacement": "general research agent",
+      "description": "Fix: general-purpose research agent description in discuss-phase-assumptions"
+    },
     {
       "pattern": "subagent_type=\"Explore\"",
       "replacement": "subagent_type=\"explore\"",
@@ -344,6 +368,16 @@
       "caseSensitive": true,
       "description": "Transform task subagent type to general"
     },
+    {
+      "pattern": "skill(skill=\"gsd:",
+      "replacement": "skill(skill=\"gsd-",
+      "description": "Fix: v1.30.0 introduced skill() calls with gsd: prefix instead of gsd- prefix"
+    },
+    {
+      "pattern": "`gsd:discuss-phase`",
+      "replacement": "`gsd-discuss-phase`",
+      "description": "Fix: standalone gsd:discuss-phase reference in autonomous.md note"
+    },
     {
       "pattern": "CLAUDE.md",
       "replacement": "AGENTS.md",
@@ -399,6 +433,31 @@
       "replacement": "get-shit-done/workflows/oc-set-profile.md",
       "description": "Use OpenCode related process instead of Claude Code's one"
     },
+    {
+      "pattern": "!`node \"$HOME/.config/opencode/get-shit-done/bin/gsd-tools.cjs\" config-set-model-profile $ARGUMENTS --raw`",
+      "replacement": "<objective>\n      Switch the model profile used by GSD agents. Controls which OpenCode model each agent uses, balancing quality vs token spend.\n\n      Routes to the set-profile workflow which handles:\n      - Argument validation (simple/smart/genius)\n      - Config file creation if missing\n      - Profile update in config.json\n      - Confirmation with model table display\n      </objective>\n\n      <execution_context>\n      @$HOME/.config/opencode/get-shit-done/workflows/oc-set-profile.md\n      </execution_context>\n\n      <process>\n      **Follow the set-profile workflow** from `@$HOME/.config/opencode/get-shit-done/workflows/oc-set-profile.md`.\n\n      The workflow handles all logic including:\n      1. Profile argument validation\n      2. Config file ensuring\n      3. Config reading and updating\n      4. Model table generation from MODEL_PROFILES\n      5. Confirmation display\n      </process> ",
+      "description": "Use OpenCode related process instead of Claude Code's one"
+    },
+    {
+      "pattern": "<purpose>",
+      "replacement": "<objective>",
+      "description": "gsd-reapply-patches.md"
+    },
+    {
+      "pattern": "</purpose>",
+      "replacement": "</objective>",
+      "description": "gsd-reapply-patches.md"
+    },
+    {
+      "pattern": "Manage parallel workstreams for concurrent milestone work.",
+      "replacement": "<objective>\nManage parallel workstreams for concurrent milestone work.\n</objective>",
+      "description": "gsd-reapply-patches.md"
+    },
+    {
+      "pattern": "Show the following output to the user verbatim, with no extra commentary:",
+      "replacement": "",
+      "description": "gsd-set-profile"
+    },
     {
       "pattern": "default: `balanced`",
       "replacement": "default: `simple`",
@@ -495,6 +554,7 @@
     "name: set-profile",
     "subagent_type=\"general-purpose\"",
     "subagent_type=\"Explore\"",
-    "CLAUDE.md"
+    "CLAUDE.md",
+    "config-set-model-profile"
   ]
 }