learnship 1.9.18 → 1.9.20

package/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "learnship",
   "description": "Agentic engineering done right — 42 structured workflows, persistent memory across sessions, integrated learning partner, and impeccable UI design system. Works with Claude Code, Windsurf, Cursor, Gemini CLI, OpenCode, and Codex.",
-  "version": "1.9.18",
+  "version": "1.9.20",
   "author": {
     "name": "Favio Vazquez",
     "email": "favio.vazquezp@gmail.com"

package/.cursor-plugin/plugin.json

@@ -2,7 +2,7 @@
   "name": "learnship",
   "displayName": "learnship",
   "description": "Agentic engineering done right — 42 structured workflows, persistent memory across sessions, integrated learning partner, and impeccable UI design system.",
-  "version": "1.9.18",
+  "version": "1.9.20",
   "logo": "assets/logo.png",
   "author": {
     "name": "Favio Vazquez",

package/SKILL.md

@@ -70,6 +70,26 @@ Learning checkpoints:
 
 The `impeccable` skill suite is always available for any UI work. Use its steering commands (`/audit`, `/critique`, `/polish`, `/colorize`, `/animate`, `/bolder`, `/quieter`, `/distill`, `/clarify`, `/optimize`, `/harden`, `/delight`, `/extract`, `/adapt`, `/onboard`, `/normalize`, `/teach-impeccable`) when reviewing or building user-facing interfaces.
 
+## Mandatory Gate — No Project, No Work
+
+**Before responding to any user message, check:**
+
+```
+Does .planning/PROJECT.md exist?
+```
+
+- **No** → The project has not been initialized. **Do NOT implement anything.** Tell the user:
+
+  > "This project hasn't been set up with learnship yet. Run `/new-project` to initialize it — that takes about 10 minutes and sets up the spec, roadmap, and phase structure before any code gets written.
+  >
+  > This is not optional: working without a spec means building the wrong thing. `/new-project` first."
+
+  Then stop. Do not offer to help with the task. Do not say "but I can also just fix it directly." Wait for the user to run `/new-project`.
+
+- **Yes** → Continue normally. Apply the workflow routing logic from `AGENTS.md`.
+
+**This gate applies to ALL messages** — bug reports, feature requests, "quick fixes", detailed specs, anything. The only exception: if the user is currently mid-ceremony in `/new-project` (i.e., they are answering your questions), their messages are workflow answers, not tasks to route.
+
 ## Key Behaviors
 
 - **Context efficiency**: Reference file paths rather than inlining file contents. Load context fresh when needed rather than carrying it forward.

package/gemini-extension.json

@@ -1,6 +1,6 @@
 {
   "name": "learnship",
-  "version": "1.9.18",
+  "version": "1.9.20",
   "description": "Agentic engineering done right — 42 structured workflows, persistent memory across sessions, integrated learning partner, and impeccable UI design system.",
   "author": "Favio Vazquez",
   "homepage": "https://faviovazquez.github.io/learnship/",

package/learnship/templates/agents.md

@@ -100,6 +100,10 @@ When a user sends a message — whether it's a vague idea, a specific bug report
 
 ### Decision tree — apply in order:
 
+**0. Is `/new-project` currently in progress?**
+
+If `.planning/PROJECT.md` does NOT exist but you are currently running `/new-project` (i.e., you have asked "What do you want to build?" and are waiting for answers, or you are in any step of the new-project ceremony): **the user's message is an answer to your workflow question, not a task to route.** Do NOT apply the routing protocol. Continue the `/new-project` ceremony from where you left off.
+
 **1. Is there a `.planning/PROJECT.md`?**
 - **No** → Stop. Tell the user: "No project found. Run `/new-project` to initialize." Do nothing else.
 - **Yes** → Continue to step 2.
@@ -127,6 +131,7 @@ Always tell the user which workflow you're about to invoke and why, then wait fo
 - User says "the login button is broken" → ❌ Don't fix it directly → ✅ Route to `/quick`
 - User says "I want to add dark mode" → ❌ Don't start implementing → ✅ Route to `discuss-phase`
 - User pastes a detailed spec → ❌ Don't treat it as a command to execute → ✅ Classify size, propose workflow, wait for yes
+- `/new-project` asked "What do you want to build?" and user replies with a detailed description → ❌ Don't treat as a task to route → ✅ It is ANSWER_1. Record it and ask Exchange 2.
 
 ---
 

package/learnship/workflows/new-project.md

@@ -10,6 +10,8 @@ Initialize a new project with full context gathering, optional research, require
 
 <!-- LEARNSHIP_PLATFORM_LABEL -->
 
+> **Routing protocol suspended.** While this workflow is running, every user message is an answer to a workflow question — not a task to route. Do NOT apply the request routing protocol until `/new-project` is fully complete and `.planning/PROJECT.md` exists.
+
 Check if `.planning/PROJECT.md` already exists:
 
 ```bash
@@ -18,6 +20,19 @@ python3 -c "import os; print('EXISTS' if os.path.exists('.planning/PROJECT.md')
 
 **If EXISTS:** Stop. Project already initialized. Use the `progress` workflow to see where you are.
 
+**Check for an existing codebase:**
+
+```bash
+python3 -c "
+import os, pathlib
+files = [p for p in pathlib.Path('.').rglob('*') if p.is_file() and not any(x in p.parts for x in ['.git', 'node_modules', '.planning', '__pycache__', '.venv'])]
+print('HAS_CODE' if len(files) > 2 else 'BLANK')
+print(f'{len(files)} files')
+"
+```
+
+**If HAS_CODE:** Note this internally as `EXISTING_CODEBASE = true`. You will scan the codebase briefly in Step 1b before questioning. Do NOT use existing code as an excuse to skip or shorten the questioning ceremony — the ceremony exists precisely because you need the user's intent, not just their code.
+
 Check if git is initialized:
 
 ```bash
@@ -39,6 +54,16 @@ Create the planning directory:
 mkdir -p .planning/research
 ```
 
+## Step 1b: Existing Codebase Scan (only if EXISTING_CODEBASE = true)
+
+If `EXISTING_CODEBASE = true`, do a quick structural scan before questioning so your follow-up questions are grounded in reality:
+
+```bash
+find . -maxdepth 3 -not -path './.git/*' -not -path './node_modules/*' -not -path './.planning/*' -not -path './__pycache__/*' -not -path './.venv/*' | sort | head -40
+```
+
+Note the tech stack, key directories, and any README content internally. Use this ONLY to ask sharper follow-up questions — never to infer the user's intent or skip ceremony steps.
+
 ## Step 2: Configuration
 
 Ask the user the following questions to configure the project. Ask them in a conversational way — not all at once, but grouped naturally.
@@ -128,6 +153,8 @@ This step is **strictly sequential**. You must complete each numbered exchange f
 Ask: **"What do you want to build?"**
 
 > 🛑 STOP. Wait for the user's answer. Do not continue until you have received it. Record their answer internally as `ANSWER_1`.
+>
+> ⚠️ **A detailed answer to Exchange 1 does NOT satisfy Exchanges 2–4.** No matter how thorough ANSWER_1 is — a full paragraph, a spec dump, a wall of requirements — it is raw material for the follow-up questions, not a replacement for them. You still MUST ask Exchanges 2, 3, and 4 before proceeding to Step 4. The purpose of follow-ups is not to extract information the user forgot to mention — it is to pressure-test, sharpen, and surface blind spots in what they already said.
 
 **Exchange 2 — First follow-up:**
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "learnship",
-  "version": "1.9.18",
+  "version": "1.9.20",
   "description": "Learn as you build. Build with intent. — A multi-platform agentic engineering system for Windsurf, Claude Code, Cursor, OpenCode, Gemini CLI, and Codex: spec-driven workflows, integrated learning, and production-grade design.",
   "keywords": [
     "agentic",