get-noesisgen-light 0.1.2 → 0.1.3

package/README.md

@@ -1,5 +1,9 @@
 # NOESISGEN Light
 
+[![npm version](https://img.shields.io/npm/v/get-noesisgen-light.svg)](https://www.npmjs.com/package/get-noesisgen-light)
+[![npm downloads](https://img.shields.io/npm/dm/get-noesisgen-light.svg)](https://www.npmjs.com/package/get-noesisgen-light)
+[![license](https://img.shields.io/npm/l/get-noesisgen-light.svg)](LICENSE)
+
 **Turn your chaos into a universe.**
 
 A personal OS for people who think in parallel. It learns who you are and how you think, not just what you ask.
@@ -134,12 +138,16 @@ NoesisGen Light includes four LaunchAgent templates you can configure to let you
 
 These are starting points. Configure them with your paths, your schedule, your rhythm.
 
+## What's your creative archetype?
+
+Take the [creative profile quiz](https://ginoftn.com/quiz) — 16 questions, 8 archetypes, bilingual FR/EN. Discover how you think and create before you build your system.
+
 ## What's next
 
 NoesisGen Light is the free, open source foundation. If you want to go further:
 
-- **NoesisGen** — [live workshop](https://ginoftn.com/workshop) ($49). Learn how to build and calibrate your full system. Includes access to the extended template repo. Optional Substack subscription ($9/month) for ongoing content and updates.
-- **NoesisPro** — [1:1 session(s)](https://ginoftn.com/pro) ($199–899). From guided installation to full architecture — built with you, calibrated to your cognitive profile. Substack Founder subscription included with Architect ($499) and Full Build ($899) tiers.
+- **[When AI Gets You](https://whenaigetsyou.substack.com)** — the NoesisGen community on Substack. Free articles on building your system + paid tiers ($19/month or $199/year) for deeper guides. The full NoesisGen template will be part of the paid tier.
+- **NoesisPro** — [1:1 sessions](https://ginoftn.com/noesispro) ($199–899). Built with you, calibrated to your cognitive profile.
 
 Both tiers are about expertise, not code. The code is the support — the value is in the calibration.
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "get-noesisgen-light",
-  "version": "0.1.2",
+  "version": "0.1.3",
   "description": "NOESIS Light — Your cognitive operating system starter kit. Install a personalized AI-powered second brain via Claude Code.",
   "type": "module",
   "bin": {

package/template/.claude/settings.local.json

@@ -6,6 +6,10 @@
       "Bash(wc *)",
       "Bash(ls *)",
       "Bash(launchctl *)",
+      "Bash(find *)",
+      "Bash(mkdir *)",
+      "Bash(chmod *)",
+      "Bash(gzip *)",
       "Read",
       "Write",
       "Edit",

package/template/.claude/skills/recap/SKILL.md

@@ -3,7 +3,7 @@ name: recap
 description: End-of-session summary. Updates daily note with what was done, decisions, and next steps.
 user-invocable: true
 disable-model-invocation: false
-argument-hint: "[write] -- no argument = display only, 'write' = write to daily note"
+argument-hint: "[display] -- no argument = display + write to daily note, 'display' = display only (no write)"
 ---
 
 # Recap -- Session summary
@@ -64,9 +64,10 @@ Generate a structured summary of the current session.
 ### `/recap` (display mode)
 - Generate and display the recap in chat
 - Touch no files
-- User can review, adjust, then ask `/recap write`
+- Default behavior: display AND write to daily note (reduces friction for ADHD users)
+- If the user wants to review first: `/recap display` shows without writing
 
-### `/recap write` (write mode)
+### `/recap display` (display-only mode)
 - Generate the recap AND append to today's daily note
 - If daily note doesn't exist, create it from template
 - Confirm with file path

package/template/.claude/skills/status/SKILL.md

@@ -57,6 +57,7 @@ Display an ASCII dashboard of the cognitive system state.
 - Sort: priority first, then by recent activity
 - If a pattern is detected (burst, silence >7d on priority, dispersion), show an alert line
 - **Agents:** check current time before marking absent. `⏳` = not yet (before scheduled time), `✓` = OK, `✗` = error or missed
+- **First-day guard:** if no log files exist yet, display `⏳ First day — agents have not run yet` instead of showing all agents as `✗`. This is normal after fresh setup.
 
 ## `/status PROJECT` (project focus)
 

package/template/CLAUDE.md

@@ -214,9 +214,17 @@ Propose a name calibrated to the conversation. Short, resonates, sounds good in
 
 Detect shell. Propose 4 aliases (`{{name}}`, `{{name}}r`, `{{name}}status`, `{{name}}brief`). Ask confirmation before writing. If refused, display for manual addition.
 
+### CRITICAL: `{{THREADS}}` generation
+
+When generating the final CLAUDE.md, you MUST populate the `{{THREADS}}` placeholder:
+- If the conversation surfaced clear themes: write 2-4 threads with one-line descriptions.
+- If the user was terse or you could not identify threads: write a placeholder:
+  `*To identify over sessions — patterns across [Project A], [Project B]. The system will detect and propose threads as it learns your work.*`
+- NEVER leave `{{THREADS}}` as a literal unfilled variable.
+
 ### Step 10: Finalization
 
-1. Read `CLAUDE-final.md.template`, replace ALL variables, write as new CLAUDE.md.
+1. Read `CLAUDE-final.md.template`, replace ALL variables (including `{{THREADS}}`), write as new CLAUDE.md.
 2. Display final summary with ASCII:
 
 ```
@@ -248,15 +256,15 @@ Detect shell. Propose 4 aliases (`{{name}}`, `{{name}}r`, `{{name}}status`, `{{n
 3. **Mirror user energy.** Brief if they're brief. Detailed if they elaborate.
 4. **One theme at a time.** Ask, wait, then move on.
 5. **NEVER react to background agent completion mid-conversation.** Continue the current theme naturally. Use scan results only at Theme 3.
-16. **ALWAYS complete the setup.** If the user provides enough information to build a profile (name, activity, at least 1 project, some sense of how they work), proceed to file generation. Do NOT keep asking questions indefinitely. A partial profile is better than no profile — it grows over sessions. If the user says "do the setup" or "fais le setup complet", that is an explicit instruction to generate NOW with what you have. Imperfect information is fine. Missing fields get "To explore in future sessions" — never block generation waiting for perfect data.
-17. **Maximum 6 questions before generating.** After 6 exchanges (not counting identity check), move to file generation regardless. The profile is enriched over time, not perfected at setup.
-6. **Cross-check identity.** If scan finds a different name than the user gave, flag it immediately.
-7. **Never read personal file contents without context.** Scan reads structure, voice analyzer reads style — not meaning.
-8. **If a step fails, continue.** Must produce a working system regardless.
-9. **Keep conversation under 15 minutes.** Profile grows over sessions.
-10. **Everything works without setup.** Skills, vault, gamification are there. Setup only personalizes.
-11. **Communicate in the user's language.** Detect from their first message. ALL generated content in their language — including CLAUDE.md final section titles, descriptions, instructions, file contents. Only code-level identifiers (USER.ENV, SHARED.ENV, skill names) stay in English.
-12. **Verify dates.** Always run `date` before writing day names.
-13. **Show progress.** Display ✓ checkmarks as files are generated. Make it feel like an installation, not a silent write.
-14. **Auto-create tasks on deadlines.** When the user mentions a deadline ("call tomorrow at 2pm", "delivery Friday"), create a task in `vault/SHARED.ENV/queue/pending.md` automatically. Confirm briefly ("Noted in your tasks.").
-15. **First session after setup.** When generating the final CLAUDE.md, include a "First session" section: if only one daily note exists (the setup one), automatically present what the system can do + available commands. Don't wait for the user to ask.
+6. **ALWAYS complete the setup.** If the user provides enough information to build a profile (name, activity, at least 1 project, some sense of how they work), proceed to file generation. Do NOT keep asking questions indefinitely. A partial profile is better than no profile — it grows over sessions. If the user says "do the setup" or "fais le setup complet", that is an explicit instruction to generate NOW with what you have. Imperfect information is fine. Missing fields get "To explore in future sessions" — never block generation waiting for perfect data.
+7. **Maximum 6 questions before generating.** After 6 exchanges (not counting identity check), move to file generation regardless. The profile is enriched over time, not perfected at setup.
+8. **Cross-check identity.** If scan finds a different name than the user gave, flag it immediately.
+9. **Never read personal file contents without context.** Scan reads structure, voice analyzer reads style — not meaning.
+10. **If a step fails, continue.** Must produce a working system regardless.
+11. **Keep conversation under 15 minutes.** Profile grows over sessions.
+12. **Everything works without setup.** Skills, vault, gamification are there. Setup only personalizes.
+13. **Communicate in the user's language.** Detect from their first message. ALL generated content in their language — including CLAUDE.md final section titles, descriptions, instructions, file contents. Only code-level identifiers (USER.ENV, SHARED.ENV, skill names) stay in English.
+14. **Verify dates.** Always run `date` before writing day names.
+15. **Show progress.** Display ✓ checkmarks as files are generated. Make it feel like an installation, not a silent write.
+16. **Auto-create tasks on deadlines.** When the user mentions a deadline ("call tomorrow at 2pm", "delivery Friday"), create a task in `vault/SHARED.ENV/queue/pending.md` automatically. Confirm briefly ("Noted in your tasks.").
+17. **First session after setup.** When generating the final CLAUDE.md, include a "First session" section: if only one daily note exists (the setup one), automatically present what the system can do + available commands. Don't wait for the user to ask.

package/template/scripts/noesis-daily-digest.sh

@@ -58,15 +58,22 @@ if [ -d "$CAPTURES_DIR" ]; then
     RECENT_CAPTURES=$(find "$CAPTURES_DIR" -name "*.md" -mtime -3 -exec basename {} \; 2>/dev/null | head -10)
 fi
 
-PROMPT=$(cat <<'PROMPT_EOF'
-You are the NOESIS daily digest agent. Generate today's brief.
-
-CONTEXT:
-- Yesterday's daily note: %YESTERDAY%
-- Active objectives: %OBJECTIVES%
-- Momentum: %MOMENTUM%
-- Pending tasks: %PENDING%
-- Recent captures: %CAPTURES%
+# Write context to temp files (avoids bash substitution bugs on multiline content)
+TMPDIR=$(mktemp -d)
+echo "$YESTERDAY_CONTEXT" > "$TMPDIR/yesterday.txt"
+echo "$OBJECTIVES_CONTEXT" > "$TMPDIR/objectives.txt"
+echo "$MOMENTUM_CONTEXT" > "$TMPDIR/momentum.txt"
+echo "$PENDING_CONTEXT" > "$TMPDIR/pending.txt"
+echo "$RECENT_CAPTURES" > "$TMPDIR/captures.txt"
+
+PROMPT="You are the NOESIS daily digest agent. Generate today's brief.
+
+Read the following context files before generating:
+- Yesterday's daily note: $TMPDIR/yesterday.txt
+- Active objectives: $TMPDIR/objectives.txt
+- Momentum: $TMPDIR/momentum.txt
+- Pending tasks: $TMPDIR/pending.txt
+- Recent captures: $TMPDIR/captures.txt
 
 INSTRUCTIONS:
 1. Read the user's profil.md and voice-dna.md Part 1 to know their language and tone.
@@ -75,19 +82,13 @@ INSTRUCTIONS:
    - **Today's focus** (based on objectives and momentum)
    - **Tasks for today** (generate 2-4 concrete tasks, add to pending.md if not already there)
    - **Signal** (any pattern worth noting — burst, stagnation, new project started)
-3. Save the digest to: vault/AI.ENV/outputs/digest-YYYY-MM-DD.md (use today's date)
+3. Save the digest to: vault/AI.ENV/outputs/digest-${DATE}.md (use today's date)
 4. Also save a copy to: vault/AI.ENV/outputs/digest-latest.md (overwrite)
 5. If there are new tasks, add them to vault/SHARED.ENV/queue/pending.md
-6. Keep it under 30 lines. Compact. Every line carries information.
-PROMPT_EOF
-)
-
-# Replace placeholders
-PROMPT="${PROMPT//%YESTERDAY%/$YESTERDAY_CONTEXT}"
-PROMPT="${PROMPT//%OBJECTIVES%/$OBJECTIVES_CONTEXT}"
-PROMPT="${PROMPT//%MOMENTUM%/$MOMENTUM_CONTEXT}"
-PROMPT="${PROMPT//%PENDING%/$PENDING_CONTEXT}"
-PROMPT="${PROMPT//%CAPTURES%/$RECENT_CAPTURES}"
+6. Keep it under 30 lines. Compact. Every line carries information."
+
+# Cleanup temp files after Claude finishes (trap ensures cleanup on error too)
+trap 'rm -rf "$TMPDIR"' EXIT
 
 # Run Claude
 cd "$(dirname "$SCRIPT_DIR")"

package/template/scripts/noesis-maintenance.sh

@@ -23,10 +23,8 @@ log "=== Starting monthly maintenance ==="
 
 # 1. Log rotation: compress logs older than 30 days
 log "Rotating logs older than 30 days..."
-ROTATED=0
-find "$LOG_DIR" -name "*.log" -mtime +30 -not -name "maintenance-*" 2>/dev/null | while read -r f; do
-    gzip "$f" 2>/dev/null && ROTATED=$((ROTATED + 1))
-done
+ROTATED=$(find "$LOG_DIR" -name "*.log" -mtime +30 -not -name "maintenance-*" 2>/dev/null | wc -l | tr -d ' ')
+find "$LOG_DIR" -name "*.log" -mtime +30 -not -name "maintenance-*" -exec gzip {} \; 2>/dev/null
 log "Rotated $ROTATED log file(s)."
 
 # Remove compressed logs older than 90 days
@@ -40,8 +38,8 @@ if [ -d "$ROOT_DIR/.git" ]; then
     cd "$ROOT_DIR"
     git add -A 2>/dev/null
     CHANGES=$(git status --porcelain 2>/dev/null | wc -l | tr -d ' ')
-    if [ "$CHANGES" -gt 0 ]; then
-        git commit -m "vault: monthly auto-commit $DATE" 2>/dev/null
+    if [ "${CHANGES:-0}" -gt 0 ]; then
+        git commit -m "vault: monthly auto-commit $DATE" 2>/dev/null || true
         log "Committed $CHANGES change(s)."
     else
         log "No changes to commit."

package/template/scripts/noesis-session-auto.sh

@@ -23,34 +23,34 @@ log() {
 
 log "=== Starting auto session ==="
 
-PROMPT=$(cat <<'PROMPT_EOF'
+PROMPT=$(cat <<PROMPT_EOF
 You are the NOESIS autonomous session agent. Work in the background.
 
 INSTRUCTIONS:
-1. Read the user's profil.md, voice-dna.md, and CLAUDE.local.md for context and rules.
+1. Read the user profil.md, voice-dna.md, and CLAUDE.local.md for context and rules.
 2. Read today's digest (vault/AI.ENV/outputs/digest-latest.md) for today's focus.
 3. Read pending tasks (vault/SHARED.ENV/queue/pending.md).
 4. Read momentum (vault/SHARED.ENV/registres/momentum.md).
 5. Read active projects (vault/SHARED.ENV/registres/projets-actifs.md).
 
-DO (pick 2-3 from this list based on what's most useful today):
-- Complete pending tasks that don't require user input
+DO (pick 2-3 from this list based on what is most useful today):
+- Complete pending tasks that do not require user input
 - Update momentum.md with current observations
 - Identify connections between active projects
 - Flag patterns (stagnation, burst activity, new directions)
-- Prepare material for the user's next session (analysis, suggestions)
+- Prepare material for the next session (analysis, suggestions)
 - Update project files if status has changed
 
 DO NOT:
 - Modify USER.ENV files (read-only)
 - Make decisions on behalf of the user
 - Create new projects or objectives
-- Use affirmative language for proposals ("this could be" not "this is")
+- Use affirmative language for proposals (this could be -- not -- this is)
 
 Write outputs to vault/AI.ENV/outputs/ with descriptive filenames.
 Update the daily note (vault/SHARED.ENV/daily-notes/YYYY-MM-DD.md) with a session summary.
 Mark completed tasks in pending.md with [x] and completion date.
-Write in the user's primary language.
+Write in the user primary language.
 Keep total output concise. Quality over quantity.
 PROMPT_EOF
 )

package/template/vault/SHARED.ENV/gamification/config.md

@@ -1,7 +1,7 @@
 # Gamification -- Configuration
 
-*Systeme de progression. Dopamine immediate, pas de punition, grace periods.*
-*Les categories XP sont adaptees au profil detecte lors du setup.*
+*Progression system. Immediate dopamine, no punishment, grace periods.*
+*XP categories are adapted to the user's profile during setup.*
 
 ---
 
@@ -9,78 +9,77 @@
 
 ### Categories generiques
 
-| Categorie | Action | XP | Condition |
-|-----------|--------|-----|-----------|
-| Ecriture | 100 mots ecrits | 15 | Tout projet creatif |
-| Ecriture | 500 mots en une session | 40 | Bonus cumul |
-| Ecriture | 1000 mots en une session | 80 | Bonus cumul |
-| Code | Commit significatif | 20 | Avancee reelle |
-| Code | Feature terminee | 80 | Feature complete |
-| Organisation | Session de travail productive | 20 | Avancee significative |
-| Organisation | Tache completee | 10 | Via /task |
-| Reflexion | Connexion identifiee | 15 | Lien entre projets |
-| Reflexion | Pattern documente | 25 | Nouveau pattern dans registre |
-| Publication | Contenu publie | 120 | Publie, pas juste programme |
-| Publication | Premiere publication | 200 | Bonus deblocage |
-| Projet | Milestone atteint | 250 | Etape majeure |
-| Projet | Projet termine | 1000 | Completion confirmee |
-| Systeme | Session interactive | 3 | Chaque session |
-
-*Ces categories sont des valeurs par defaut. Le setup les adapte selon les projets detectes.*
-*Tu peux les modifier a tout moment.*
+| Category | Action | XP | Condition |
+|----------|--------|-----|-----------|
+| Writing | 100 words written | 15 | Any creative project |
+| Writing | 500 words in one session | 40 | Cumulative bonus |
+| Writing | 1000 words in one session | 80 | Cumulative bonus |
+| Code | Significant commit | 20 | Real progress |
+| Code | Feature finished | 80 | Complete feature |
+| Organization | Productive work session | 20 | Significant progress |
+| Organization | Task completed | 10 | Via /task |
+| Reflection | Connection identified | 15 | Link between projects |
+| Reflection | Pattern documented | 25 | New pattern in register |
+| Publication | Content published | 120 | Published, not just scheduled |
+| Publication | First publication | 200 | First-time bonus |
+| Project | Milestone reached | 250 | Major milestone |
+| Project | Project completed | 1000 | Confirmed completion |
+| System | Interactive session | 3 | Each session |
+
+*These are default categories. Setup adapts them to detected projects. You can modify them anytime.*
 
 ---
 
 ## Streaks
 
-### Streak principal
-- **Condition :** Activite creative ou productive (100 mots, 1 commit, 1 tache)
-- **Grace period :** 48h (2 jours sans activite avant que le streak se pause)
-- **Bonus :** +5 XP par jour de streak actif (plafond : +50 XP/jour a 10 jours)
-- **Pause vs bris :** Le streak se **pause** (compteur gele), il ne se **brise** jamais. Reprendre remet le compteur ou il etait.
+### Main streak
+- **Condition:** Creative or productive activity (100 words, 1 commit, 1 task)
+- **Grace period:** 48h (2 days without activity before the streak pauses)
+- **Bonus:** +5 XP per active streak day (cap: +50 XP/day at 10 days)
+- **Pause vs break:** The streak **pauses** (counter frozen), it never **breaks**. Resuming restores the counter where it was.
 
-### Regle anti-pression
-- Les sessions autonomes NE MENTIONNENT JAMAIS le streak
-- Le streak est une info passive consultable via /status, jamais affichee proactivement
-- **Celebrer, jamais culpabiliser.**
+### Anti-pressure rule
+- Autonomous sessions NEVER mention the streak
+- The streak is passive info viewable via /status, never displayed proactively
+- **Celebrate, never guilt.**
 
 ---
 
-## Niveaux
-
-18 niveaux en 3 actes. Les premiers tombent vite (dopamine). Les derniers demandent un vrai effort.
-
-### Acte I : L'Eveil (niveaux 1-5)
-| Niveau | Nom | XP requis |
-|--------|-----|-----------|
-| 1 | Le Dormeur | 0 |
-| 2 | L'Eveille | 100 |
-| 3 | Le Curieux | 300 |
-| 4 | L'Apprenti | 600 |
-| 5 | L'Initie | 1000 |
-
-### Acte II : La Forge (niveaux 6-12)
-| Niveau | Nom | XP requis |
-|--------|-----|-----------|
-| 6 | L'Artisan | 1500 |
-| 7 | Le Forgeur | 2200 |
-| 8 | L'Obsede | 3000 |
-| 9 | Le Marathonien | 4000 |
-| 10 | Le Fidele | 5200 |
-| 11 | L'Enracine | 6500 |
-| 12 | Le Constructeur | 8000 |
-
-### Acte III : La Transmission (niveaux 13-18)
-| Niveau | Nom | XP requis |
-|--------|-----|-----------|
-| 13 | L'Incandescent | 10000 |
-| 14 | Le Passeur | 12500 |
-| 15 | L'Architecte | 15000 |
-| 16 | Le Visionnaire | 18000 |
-| 17 | Le Sage | 21500 |
-| 18 | Le Maitre | 25000 |
-
-*Les noms de niveaux sont personnalisables. Le setup peut proposer des noms adaptes a l'univers de l'utilisateur.*
+## Levels
+
+18 levels in 3 acts. Early levels come fast (dopamine). Later ones require real effort.
+
+### Act 1 — Awakening (lv. 1-5)
+| Level | Name | XP required |
+|-------|------|-------------|
+| 1 | Sleeper | 0 |
+| 2 | Spark | 100 |
+| 3 | Explorer | 300 |
+| 4 | Apprentice | 600 |
+| 5 | Initiate | 1000 |
+
+### Act 2 — Building (lv. 6-12)
+| Level | Name | XP required |
+|-------|------|-------------|
+| 6 | Maker | 1500 |
+| 7 | Builder | 2200 |
+| 8 | Shaper | 3000 |
+| 9 | Architect | 4000 |
+| 10 | Adept | 5200 |
+| 11 | Mentor | 6500 |
+| 12 | Guide | 8000 |
+
+### Act 3 — Mastery (lv. 13-18)
+| Level | Name | XP required |
+|-------|------|-------------|
+| 13 | Pioneer | 10000 |
+| 14 | Weaver | 12500 |
+| 15 | Strategist | 15000 |
+| 16 | Guardian | 18000 |
+| 17 | Founder | 21500 |
+| 18 | Sovereign | 25000 |
+
+*Level names are customizable. The setup agent proposes names adapted to the user's world.*
 
 ---