worclaude 2.2.0 → 2.2.3

package/README.md

@@ -48,9 +48,9 @@ Worclaude scaffolds a complete Claude Code workflow into any project in seconds.
 
 ---
 
-## What's New in v2.0.0
+## What's New in v2.x
 
-v2.0.0 is a major integration release. Skills and agents now register with Claude Code's runtime systems — they show up in `/skills` and `/agents`, not just as files on disk.
+The v2.x series introduced Claude Code runtime integration and continues to refine the workflow.
 
 **Claude Code Runtime Integration**
 
@@ -74,7 +74,7 @@ v2.0.0 is a major integration release. Skills and agents now register with Claud
 **New Content**
 
 - Coordinator-mode skill for multi-agent orchestration patterns
-- Optional MEMORY.md template for Claude Code's memory system
+- Agents with `memory: project` leverage Claude Code's native memory system
 - Enhanced verify-app agent with structured verdicts and adversarial probe requirements
 - Per-tech-stack permission presets (16 languages)
 
@@ -83,6 +83,12 @@ v2.0.0 is a major integration release. Skills and agents now register with Claud
 - `worclaude upgrade` auto-migrates v1.x projects: flat skills → directory format, agents get required frontmatter
 - `worclaude doctor` detects old formats and missing fields
 
+**v2.2.0 — Leaner Defaults, Richer Skills**
+
+- Removed project-root MEMORY.md scaffolding — Claude Code's native memory system handles this automatically
+- Expanded context-management, claude-md-maintenance, and coordinator-mode skills with deeper guidance
+- Enhanced verify-app agent with mobile, database migration, and data/ML pipeline verification patterns
+
 See the [Claude Code Integration guide](https://sefaertunc.github.io/Worclaude/guide/claude-code-integration) for technical details.
 
 ---

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "worclaude",
-  "version": "2.2.0",
+  "version": "2.2.3",
   "description": "CLI tool that scaffolds a comprehensive Claude Code workflow into any project",
   "type": "module",
   "bin": {

package/src/commands/init.js

@@ -211,27 +211,11 @@ async function runAgents(selections) {
   return { ...selections, selectedAgents };
 }
 
-async function runMemoryMd(selections) {
-  const { includeMemoryMd } = await inquirer.prompt([
-    {
-      type: 'list',
-      name: 'includeMemoryMd',
-      message: 'Scaffold a MEMORY.md template? (for Claude Code memory system)',
-      choices: [
-        { name: 'No', value: false },
-        { name: 'Yes', value: true },
-      ],
-    },
-  ]);
-  return { ...selections, includeMemoryMd };
-}
-
 const STEP_RUNNERS = {
   projectInfo: runProjectInfo,
   projectType: runProjectType,
   techStack: runTechStack,
   agents: runAgents,
-  memoryMd: runMemoryMd,
 };
 
 // --- Confirmation ---
@@ -295,7 +279,6 @@ async function runInteractivePrompts(projectRoot) {
     languages: [],
     useDocker: false,
     selectedAgents: [],
-    includeMemoryMd: false,
   };
 
   let confirmed = false;
@@ -307,7 +290,6 @@ async function runInteractivePrompts(projectRoot) {
       selections = await runProjectType(selections);
       selections = await runTechStack(selections);
       selections = await runAgents(selections);
-      selections = await runMemoryMd(selections);
       firstRun = false;
     }
 
@@ -323,7 +305,6 @@ async function runInteractivePrompts(projectRoot) {
         languages: [],
         useDocker: false,
         selectedAgents: [],
-        includeMemoryMd: false,
       };
       display.newline();
       display.info('Starting over...');
@@ -332,7 +313,6 @@ async function runInteractivePrompts(projectRoot) {
       selections = await runProjectType(selections);
       selections = await runTechStack(selections);
       selections = await runAgents(selections);
-      selections = await runMemoryMd(selections);
     } else if (confirmation === 'adjust') {
       const { step } = await inquirer.prompt([
         {
@@ -520,11 +500,6 @@ async function scaffoldFresh(projectRoot, selections, variables, settingsStr, ve
     await writeFile(path.join(projectRoot, '.claude', 'sessions', '.gitkeep'), '');
     spinner.text = 'Created .claude/sessions/';
 
-    if (selections.includeMemoryMd) {
-      await scaffoldFile('core/memory-md.md', 'MEMORY.md', {}, projectRoot);
-      spinner.text = 'Created MEMORY.md';
-    }
-
     await computeAndWriteWorkflowMeta(projectRoot, selections, version);
     spinner.text = 'Created .claude/workflow-meta.json';
 
@@ -551,9 +526,6 @@ function displayFreshSuccess(selections, skipped) {
   display.success('.claude/sessions/');
   display.success('.mcp.json');
   display.success('.gitignore');
-  if (selections.includeMemoryMd) {
-    display.success('MEMORY.md');
-  }
   if (skipped.progressMd) {
     display.dim('  docs/spec/PROGRESS.md — already exists, skipped');
   }

package/src/data/agents.js

@@ -315,7 +315,6 @@ export const CONFIRMATION_STEPS = [
   { name: 'Project type', value: 'projectType' },
   { name: 'Tech stack', value: 'techStack' },
   { name: 'Agent selection', value: 'agents' },
-  { name: 'MEMORY.md (optional)', value: 'memoryMd' },
 ];
 
 export const NOTIFICATION_COMMANDS = {

package/templates/skills/universal/context-management.md

@@ -133,7 +133,7 @@ Approximate values from Claude Code v2.1.88 source (March 2026). These may chang
 | Post-compact skills restore | 25,000 token budget, 5,000 per skill |
 | Compact summary output | 20,000 tokens max |
 | CLAUDE.md total budget | ~40,000 characters across all loaded instruction files |
-| MEMORY.md | 200 lines / 25,000 bytes |
+| MEMORY.md (native, ~/.claude/) | 200 lines / 25,000 bytes |
 
 Practical takeaways:
 - After compaction, only 5 files are restored at 5k tokens each. Structure your work

package/templates/core/memory-md.md

@@ -1,70 +0,0 @@
-# MEMORY.md
-
-> This file is an index of pointers, not storage. Each entry should be under ~200 characters.
-> Limit: 200 lines / 25KB. Move detail into topic files in `.claude/memory/`.
-> Claude Code reads this at session start — keep it lean.
-
-## User
-
-<!-- Role, preferences, expertise — helps Claude tailor responses -->
-<!-- Example: - [Backend lead](user_role.md) — 8 years Python, new to this frontend -->
-
-## Feedback
-
-<!-- What to do and what to avoid — both corrections AND confirmed approaches -->
-<!-- Format: rule, then **Why:** and **How to apply:** -->
-<!-- Example: - [No mocks in integration tests](feedback_testing.md) — burned by mock/prod divergence -->
-
-## Project
-
-<!-- Ongoing work, goals, deadlines — convert relative dates to absolute -->
-<!-- Format: fact/decision, then **Why:** and **How to apply:** -->
-<!-- Example: - [Merge freeze 2026-03-05](project_release.md) — mobile team cutting release branch -->
-
-## Reference
-
-<!-- Pointers to external systems — Linear boards, Slack channels, dashboards -->
-<!-- Example: - [Pipeline bugs](reference_linear.md) — tracked in Linear project "INGEST" -->
-
----
-
-## Memory File Format
-
-Each memory file in `.claude/memory/` uses this frontmatter:
-
-```markdown
----
-name: descriptive-name
-description: one-line summary used to decide relevance in future sessions
-type: user | feedback | project | reference
----
-
-Content here. For feedback/project types, structure as:
-Rule or fact.
-**Why:** the reason or context.
-**How to apply:** when and where this guidance applies.
-```
-
-## Drift and Verification
-
-- Memory records become stale over time. Before acting on a memory, verify it against the current state of the codebase.
-- If a memory names a file path, check the file still exists. If it names a function or flag, grep for it.
-- If a recalled memory conflicts with current information, trust what you observe now — update or remove the stale memory.
-- "The memory says X exists" is not the same as "X exists now."
-
-## What NOT to save here
-
-- Code patterns, architecture, file paths — derivable from current project state
-- Git history, recent changes — `git log` / `git blame` are authoritative
-- Debugging solutions — the fix is in the code, the context is in the commit message
-- Anything already in CLAUDE.md
-- Ephemeral task details — current conversation context, in-progress work
-
-These exclusions apply even when explicitly asked to save. If asked to save a PR list or activity summary, ask what was *surprising* or *non-obvious* about it — that is the part worth keeping.
-
-## Memory vs Plans vs Tasks
-
-Memory is for information useful across conversations. Don't use it for:
-- **Plans** — use a plan file or plan mode for implementation strategies within a session
-- **Tasks** — use the task system for tracking work steps within a session
-- **Session state** — current conversation context belongs in the conversation, not memory