@askexenow/exe-os 0.8.28 → 0.8.30

package/dist/bin/cli.js

@@ -4524,13 +4524,24 @@ You are \${agent_id}. COO \u2014 the founder's most reliable teammate in busines
 - Pattern recognition: surface recurring problems, connect dots across projects
 - Architecture guardian (strategic): verify all work aligns with the PRODUCT VISION and five-mode architecture in .planning/ARCHITECTURE.md. Is this the right feature at the right time? Does it match the build order?
 
-## First Boot \u2014 Memory Onboarding
-
-On your FIRST conversation (when you have few or no prior interactions with the founder):
-1. Run recall_my_memory with broad queries to discover what exists: search for "project", "client", "pipeline", "SEO", "campaign", "product", "deal", "meeting"
-2. Summarize what you find: "You've been working on X, Y, Z. Here's what I know about each."
-3. If memories exist from a backfill, proactively tell the founder what you learned from their history
-4. Ask what's most important right now
+## Every Session \u2014 Status Brief
+
+On EVERY new conversation, before doing anything else:
+
+1. **Memory scan**: Run recall_my_memory with broad queries \u2014 "project", "client", "pipeline", "campaign", "deal", "decision", "blocker". Summarize what you find.
+2. **Task scan**: Run list_tasks to see what's open, in progress, blocked, or needs review across all employees.
+3. **Team check**: Run ask_team_memory for recent activity from CTO/CMO/engineers.
+4. **Present the brief**: Give the founder a concise status report:
+   - What's active and progressing
+   - What's blocked and needs attention
+   - What decisions are pending
+   - What you recommend doing next
+5. Then ask: "What's the priority?"
+
+If this is your FIRST ever conversation (few or no prior memories):
+- Search more broadly: "product", "SEO", "meeting", "strategy", "revenue"
+- Proactively summarize what you learned from backfilled history
+- Introduce yourself and what you can do
 
 Never say "I have no memories" without first searching broadly. Your memory may contain thousands of entries \u2014 surface them by searching for domain-relevant terms, not meta-queries like "what do I know."
 

package/dist/bin/exe-new-employee.js

@@ -196,13 +196,24 @@ You are \${agent_id}. COO \u2014 the founder's most reliable teammate in busines
 - Pattern recognition: surface recurring problems, connect dots across projects
 - Architecture guardian (strategic): verify all work aligns with the PRODUCT VISION and five-mode architecture in .planning/ARCHITECTURE.md. Is this the right feature at the right time? Does it match the build order?
 
-## First Boot \u2014 Memory Onboarding
-
-On your FIRST conversation (when you have few or no prior interactions with the founder):
-1. Run recall_my_memory with broad queries to discover what exists: search for "project", "client", "pipeline", "SEO", "campaign", "product", "deal", "meeting"
-2. Summarize what you find: "You've been working on X, Y, Z. Here's what I know about each."
-3. If memories exist from a backfill, proactively tell the founder what you learned from their history
-4. Ask what's most important right now
+## Every Session \u2014 Status Brief
+
+On EVERY new conversation, before doing anything else:
+
+1. **Memory scan**: Run recall_my_memory with broad queries \u2014 "project", "client", "pipeline", "campaign", "deal", "decision", "blocker". Summarize what you find.
+2. **Task scan**: Run list_tasks to see what's open, in progress, blocked, or needs review across all employees.
+3. **Team check**: Run ask_team_memory for recent activity from CTO/CMO/engineers.
+4. **Present the brief**: Give the founder a concise status report:
+   - What's active and progressing
+   - What's blocked and needs attention
+   - What decisions are pending
+   - What you recommend doing next
+5. Then ask: "What's the priority?"
+
+If this is your FIRST ever conversation (few or no prior memories):
+- Search more broadly: "product", "SEO", "meeting", "strategy", "revenue"
+- Proactively summarize what you learned from backfilled history
+- Introduce yourself and what you can do
 
 Never say "I have no memories" without first searching broadly. Your memory may contain thousands of entries \u2014 surface them by searching for domain-relevant terms, not meta-queries like "what do I know."
 

package/dist/bin/setup.js

@@ -1871,13 +1871,24 @@ You are \${agent_id}. COO \u2014 the founder's most reliable teammate in busines
 - Pattern recognition: surface recurring problems, connect dots across projects
 - Architecture guardian (strategic): verify all work aligns with the PRODUCT VISION and five-mode architecture in .planning/ARCHITECTURE.md. Is this the right feature at the right time? Does it match the build order?
 
-## First Boot \u2014 Memory Onboarding
-
-On your FIRST conversation (when you have few or no prior interactions with the founder):
-1. Run recall_my_memory with broad queries to discover what exists: search for "project", "client", "pipeline", "SEO", "campaign", "product", "deal", "meeting"
-2. Summarize what you find: "You've been working on X, Y, Z. Here's what I know about each."
-3. If memories exist from a backfill, proactively tell the founder what you learned from their history
-4. Ask what's most important right now
+## Every Session \u2014 Status Brief
+
+On EVERY new conversation, before doing anything else:
+
+1. **Memory scan**: Run recall_my_memory with broad queries \u2014 "project", "client", "pipeline", "campaign", "deal", "decision", "blocker". Summarize what you find.
+2. **Task scan**: Run list_tasks to see what's open, in progress, blocked, or needs review across all employees.
+3. **Team check**: Run ask_team_memory for recent activity from CTO/CMO/engineers.
+4. **Present the brief**: Give the founder a concise status report:
+   - What's active and progressing
+   - What's blocked and needs attention
+   - What decisions are pending
+   - What you recommend doing next
+5. Then ask: "What's the priority?"
+
+If this is your FIRST ever conversation (few or no prior memories):
+- Search more broadly: "product", "SEO", "meeting", "strategy", "revenue"
+- Proactively summarize what you learned from backfilled history
+- Introduce yourself and what you can do
 
 Never say "I have no memories" without first searching broadly. Your memory may contain thousands of entries \u2014 surface them by searching for domain-relevant terms, not meta-queries like "what do I know."
 

package/dist/lib/identity-templates.js

@@ -47,13 +47,24 @@ You are \${agent_id}. COO \u2014 the founder's most reliable teammate in busines
 - Pattern recognition: surface recurring problems, connect dots across projects
 - Architecture guardian (strategic): verify all work aligns with the PRODUCT VISION and five-mode architecture in .planning/ARCHITECTURE.md. Is this the right feature at the right time? Does it match the build order?
 
-## First Boot \u2014 Memory Onboarding
-
-On your FIRST conversation (when you have few or no prior interactions with the founder):
-1. Run recall_my_memory with broad queries to discover what exists: search for "project", "client", "pipeline", "SEO", "campaign", "product", "deal", "meeting"
-2. Summarize what you find: "You've been working on X, Y, Z. Here's what I know about each."
-3. If memories exist from a backfill, proactively tell the founder what you learned from their history
-4. Ask what's most important right now
+## Every Session \u2014 Status Brief
+
+On EVERY new conversation, before doing anything else:
+
+1. **Memory scan**: Run recall_my_memory with broad queries \u2014 "project", "client", "pipeline", "campaign", "deal", "decision", "blocker". Summarize what you find.
+2. **Task scan**: Run list_tasks to see what's open, in progress, blocked, or needs review across all employees.
+3. **Team check**: Run ask_team_memory for recent activity from CTO/CMO/engineers.
+4. **Present the brief**: Give the founder a concise status report:
+   - What's active and progressing
+   - What's blocked and needs attention
+   - What decisions are pending
+   - What you recommend doing next
+5. Then ask: "What's the priority?"
+
+If this is your FIRST ever conversation (few or no prior memories):
+- Search more broadly: "product", "SEO", "meeting", "strategy", "revenue"
+- Proactively summarize what you learned from backfilled history
+- Introduce yourself and what you can do
 
 Never say "I have no memories" without first searching broadly. Your memory may contain thousands of entries \u2014 surface them by searching for domain-relevant terms, not meta-queries like "what do I know."
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@askexenow/exe-os",
-  "version": "0.8.28",
+  "version": "0.8.30",
   "description": "AI employee operating system — persistent memory, task management, and multi-agent coordination for Claude Code.",
   "license": "CC-BY-NC-4.0",
   "type": "module",

package/src/commands/exe/assign.md

@@ -8,10 +8,10 @@ Assign a task to the best specialist or a named employee.
 
 Auto-route (finds best match):
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-assign.js" "debug the auth timeout"
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-assign.js" "debug the auth timeout"
 ```
 
 Direct assign:
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-assign.js" yoshi "debug the auth timeout"
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-assign.js" yoshi "debug the auth timeout"
 ```

package/src/commands/exe/build-adv.md

@@ -181,8 +181,8 @@ For each subtask, call create_task with:
 **VERIFY:**
 ```bash
 SUBTASK_COUNT=$(node -e "
-  const {getClient}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'exe-os/dist/lib/turso.js'));
-  const {initStore}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'exe-os/dist/lib/store.js'));
+  const {getClient}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'@askexenow/exe-os/dist/lib/turso.js'));
+  const {initStore}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'@askexenow/exe-os/dist/lib/store.js'));
   (async()=>{await initStore();const c=getClient();const r=await c.execute({sql:\"SELECT COUNT(*) as cnt FROM tasks WHERE title LIKE 'build-adv/{slug}%' AND status != 'done'\",args:[]});console.log(Number(r.rows[0]?.cnt??0))})()
 " 2>/dev/null)
 [ "$SUBTASK_COUNT" -gt 0 ] 2>/dev/null && echo "CHECKPOINT_PASS:${SUBTASK_COUNT}_subtasks" || echo "CHECKPOINT_FAIL:no_subtasks_in_db"
@@ -319,8 +319,8 @@ node -e "const f='exe/output/{slug}/build-state.json'; const s=JSON.parse(requir
 **VERIFY:**
 ```bash
 OPEN_COUNT=$(node -e "
-  const {getClient}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'exe-os/dist/lib/turso.js'));
-  const {initStore}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'exe-os/dist/lib/store.js'));
+  const {getClient}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'@askexenow/exe-os/dist/lib/turso.js'));
+  const {initStore}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'@askexenow/exe-os/dist/lib/store.js'));
   (async()=>{await initStore();const c=getClient();const r=await c.execute({sql:\"SELECT COUNT(*) as cnt FROM tasks WHERE title LIKE 'build-adv/{slug}%' AND status NOT IN ('done','cancelled')\",args:[]});console.log(Number(r.rows[0]?.cnt??0))})()
 " 2>/dev/null)
 [ "$OPEN_COUNT" -eq 0 ] 2>/dev/null && echo "CHECKPOINT_PASS" || echo "CHECKPOINT_FAIL:${OPEN_COUNT}_subtasks_still_open"

package/src/commands/exe/call.md

@@ -67,7 +67,7 @@ Employee sessions are named `{name}-{exeSession}` (e.g., `yoshi-exe1`) or `{name
 ```bash
 NAME="$ARGUMENTS"
 EXE_SESSION=$(tmux display-message -p '#{session_name}' 2>/dev/null)
-RESULT=$(node -e "const {ensureEmployee}=require('$(npm root -g)/exe-os/dist/lib/tmux-routing.js'); const r=ensureEmployee('${NAME}','${EXE_SESSION}','$(pwd)'); console.log(JSON.stringify(r));")
+RESULT=$(node -e "const {ensureEmployee}=require('$(npm root -g)/@askexenow/exe-os/dist/lib/tmux-routing.js'); const r=ensureEmployee('${NAME}','${EXE_SESSION}','$(pwd)'); console.log(JSON.stringify(r));")
 echo "$RESULT"
 ```
 Report the result and STOP HERE. Do not continue to step 1.
@@ -102,7 +102,7 @@ export AGENT_ROLE="${ROLE}"
 MY_SESSION=$(tmux display-message -p '#{session_name}' 2>/dev/null)
 DISPATCH_INFO="$HOME/.exe-os/session-cache/dispatch-info-${MY_SESSION}.json"
 DISPATCHED_BY=$(node -e "try{const d=JSON.parse(require('fs').readFileSync('$DISPATCH_INFO','utf8'));console.log(d.dispatchedBy||'')}catch{console.log('')}" 2>/dev/null)
-node -e "const {parseParentExe,registerParentExe}=require('$(npm root -g)/exe-os/dist/lib/tmux-routing.js'); const p=parseParentExe('$MY_SESSION','$NAME'); if(p) registerParentExe('$KEY',p,'$DISPATCHED_BY'||undefined);"
+node -e "const {parseParentExe,registerParentExe}=require('$(npm root -g)/@askexenow/exe-os/dist/lib/tmux-routing.js'); const p=parseParentExe('$MY_SESSION','$NAME'); if(p) registerParentExe('$KEY',p,'$DISPATCHED_BY'||undefined);"
 ```
 
 2. Ensure exe folder structure exists and scan for open tasks:
@@ -114,7 +114,7 @@ if [ -f .gitignore ]; then
 else
   echo -e '# Employee task assignments (private)\n/exe/' > .gitignore
 fi
-node "$(npm root -g)/exe-os/dist/bin/scan-tasks.js" "exe/$ARGUMENTS" --format=mandatory 2>/dev/null
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/scan-tasks.js" "exe/$ARGUMENTS" --format=mandatory 2>/dev/null
 ```
 
 3. Check memories for context from exe:

package/src/commands/exe/cloud.md

@@ -11,7 +11,7 @@ One command to link a device and enable cloud sync.
 - `/exe-cloud status` — show current sync status and key fingerprint
 
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-cloud.js" $ARGUMENTS
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-cloud.js" $ARGUMENTS
 ```
 
 **Note:** `/exe-link` is deprecated. Use `/exe-cloud` instead.

package/src/commands/exe/doctor.md

@@ -6,5 +6,5 @@ allowed-tools: Bash
 Run a full memory health audit showing total memories, per-agent breakdown, date range, and any issues.
 
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-doctor.js" $ARGUMENTS
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-doctor.js" $ARGUMENTS
 ```

package/src/commands/exe/employee-heartbeat.md

@@ -37,7 +37,7 @@ if [ -f "$ACTIVITY_FILE" ]; then
 fi
 
 # Use canonical scanner — single source of truth for task status detection
-node "$(npm root -g)/exe-os/dist/bin/scan-tasks.js" "exe/$AGENT" --format=mandatory 2>/dev/null
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/scan-tasks.js" "exe/$AGENT" --format=mandatory 2>/dev/null
 ```
 
 **If the script outputs task info:** read the task file and begin working immediately. Do not ask what to do.

package/src/commands/exe/forget.md

@@ -11,5 +11,5 @@ Usage:
 - `/exe-forget --query "<search>"` — delete memories matching a search query
 
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-forget.js" "$ARGUMENTS"
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-forget.js" "$ARGUMENTS"
 ```

package/src/commands/exe/intercom.md

@@ -24,23 +24,23 @@ fi
 
 # Exe gets review notifications instead of task dispatch
 if [ "$AGENT" = "exe" ]; then
-  node "$(npm root -g)/exe-os/dist/bin/exe-pending-reviews.js" 2>/dev/null
+  node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-pending-reviews.js" 2>/dev/null
   # Exe also subscribes to notifications (subtasks_complete, task_complete, etc.)
   # so parent-of-parent rollups surface live instead of only at boot.
-  node "$(npm root -g)/exe-os/dist/bin/exe-pending-notifications.js" "$AGENT" 2>/dev/null
+  node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-pending-notifications.js" "$AGENT" 2>/dev/null
   exit 0
 fi
 
 # Use canonical scanner — single source of truth for task status detection
-node "$(npm root -g)/exe-os/dist/bin/scan-tasks.js" "exe/$AGENT" --format=mandatory 2>/dev/null
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/scan-tasks.js" "exe/$AGENT" --format=mandatory 2>/dev/null
 
 # Check for pending messages
-node "$(npm root -g)/exe-os/dist/bin/exe-pending-messages.js" "$AGENT" 2>/dev/null
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-pending-messages.js" "$AGENT" 2>/dev/null
 
 # Check for pending notifications (subtasks_complete, rollup signals, etc.)
 # Employees acting as parent-of-parents (e.g. yoshi holding multiple parent
 # tasks) need these too — not only exe. See the parent-cascade audit §6.
-node "$(npm root -g)/exe-os/dist/bin/exe-pending-notifications.js" "$AGENT" 2>/dev/null
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-pending-notifications.js" "$AGENT" 2>/dev/null
 ```
 
 ### For exe (AGENT=exe)

package/src/commands/exe/launch.md

@@ -33,7 +33,7 @@ For each name in $ARGUMENTS, verify it exists in the roster. If any name is not
 ```bash
 NAME="<employee>"
 EXE_SESSION=$(tmux display-message -p '#{session_name}' 2>/dev/null)
-RESULT=$(node -e "const {ensureEmployee}=require('$(npm root -g)/exe-os/dist/lib/tmux-routing.js'); const r=ensureEmployee('${NAME}','${EXE_SESSION}','$(pwd)'); console.log(JSON.stringify(r));")
+RESULT=$(node -e "const {ensureEmployee}=require('$(npm root -g)/@askexenow/exe-os/dist/lib/tmux-routing.js'); const r=ensureEmployee('${NAME}','${EXE_SESSION}','$(pwd)'); console.log(JSON.stringify(r));")
 echo "${NAME}: ${RESULT}"
 ```
 

package/src/commands/exe/link.md

@@ -13,5 +13,5 @@ argument-hint: [export|import]
 Legacy command still works:
 
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-link.js" $ARGUMENTS
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-link.js" $ARGUMENTS
 ```

package/src/commands/exe/new-employee.md

@@ -8,5 +8,5 @@ Create a new named employee for the exe-os system.
 
 Run this command:
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-new-employee.js" $ARGUMENTS
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-new-employee.js" $ARGUMENTS
 ```

package/src/commands/exe/review.md

@@ -10,5 +10,5 @@ update organizational memory with the outcomes.
 
 Run this command:
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-review.js" $ARGUMENTS
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-review.js" $ARGUMENTS
 ```

package/src/commands/exe/schedule.md

@@ -21,8 +21,8 @@ Determine the action:
 
 ```bash
 node -e "
-  const {initStore}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'exe-os/dist/lib/store.js'));
-  const {listSchedules}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'exe-os/dist/lib/schedules.js'));
+  const {initStore}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'@askexenow/exe-os/dist/lib/store.js'));
+  const {listSchedules}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'@askexenow/exe-os/dist/lib/schedules.js'));
   (async()=>{
     await initStore();
     const schedules=await listSchedules();
@@ -45,8 +45,8 @@ Print the output as-is.
 ```bash
 ID="<extracted-id>"
 node -e "
-  const {initStore}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'exe-os/dist/lib/store.js'));
-  const {deleteSchedule}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'exe-os/dist/lib/schedules.js'));
+  const {initStore}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'@askexenow/exe-os/dist/lib/store.js'));
+  const {deleteSchedule}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'@askexenow/exe-os/dist/lib/schedules.js'));
   (async()=>{
     await initStore();
     const ok=await deleteSchedule('${ID}');
@@ -66,7 +66,7 @@ node -e "
 ```bash
 TIME_EXPR="<extracted-time>"
 node -e "
-  const {parseHumanCron}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'exe-os/dist/lib/schedules.js'));
+  const {parseHumanCron}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'@askexenow/exe-os/dist/lib/schedules.js'));
   console.log(parseHumanCron('${TIME_EXPR}'))
 " 2>/dev/null
 ```
@@ -89,8 +89,8 @@ DESC="<description>"
 TYPE="<job-type>"
 CRONTAB="<0-or-1>"
 node -e "
-  const {initStore}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'exe-os/dist/lib/store.js'));
-  const {createSchedule}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'exe-os/dist/lib/schedules.js'));
+  const {initStore}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'@askexenow/exe-os/dist/lib/store.js'));
+  const {createSchedule}=require(require('path').join(require('child_process').execSync('npm root -g',{encoding:'utf8'}).trim(),'@askexenow/exe-os/dist/lib/schedules.js'));
   (async()=>{
     await initStore();
     const s=await createSchedule({cron:'${CRON}',description:'${DESC}',jobType:'${TYPE}',useCrontab:${CRONTAB}===1});

package/src/commands/exe/search.md

@@ -9,5 +9,5 @@ Usage: `/exe-search <query>`
 
 Run this command with the user's query:
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-search.js" "$ARGUMENTS"
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-search.js" "$ARGUMENTS"
 ```

package/src/commands/exe/settings.md

@@ -9,5 +9,5 @@ Configure auto-ingestion, auto-retrieval, and search mode.
 Run this command to open the settings menu:
 
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-settings.js"
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-settings.js"
 ```

package/src/commands/exe/status.md

@@ -8,7 +8,7 @@ Show real-time employee status by capturing tmux pane output.
 
 Run:
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-status.js" $ARGUMENTS 2>/dev/null
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-status.js" $ARGUMENTS 2>/dev/null
 ```
 
 Print the output EXACTLY as-is — it is pre-formatted with box-drawing characters.

package/src/commands/exe/team.md

@@ -7,5 +7,5 @@ Show all registered employees, their roles, memory counts, and creation dates.
 
 Run this command:
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-team.js"
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-team.js"
 ```

package/src/commands/exe/update.md

@@ -7,5 +7,5 @@ Check if a newer version of exe-os is available on npm and optionally install it
 
 Run this command:
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/update.js"
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/update.js"
 ```

package/src/commands/exe.md

@@ -133,7 +133,7 @@ When an employee sends `context-full: <name> hit capacity. Checkpoint saved. Res
 After completing a review (or batch of reviews), re-check for new reviews:
 
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-pending-reviews.js" 2>/dev/null
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-pending-reviews.js" 2>/dev/null
 ```
 
 If more reviews exist, start reviewing them immediately. Do NOT stop and wait for the founder. Keep chaining until the queue is empty. Only then resume normal conversation.
@@ -149,7 +149,7 @@ You manage 10-20+ projects. When the founder asks "what's going on," give the re
 Run the boot script:
 
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-boot.js"
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-boot.js"
 ```
 
 **The user already sees the bash output. Do NOT reproduce, summarize, reformat, or comment on it.**
@@ -160,7 +160,7 @@ Say nothing after the script runs. The user will speak first.
 When the founder asks about status, reviews, tasks, progress, or "what's going on" at ANY point AFTER the initial boot — re-run the status query. NEVER reference the boot snapshot for current state.
 
 ```bash
-node "$(npm root -g)/exe-os/dist/bin/exe-boot.js" --brief-only
+node "$(npm root -g)/@askexenow/exe-os/dist/bin/exe-boot.js" --brief-only
 ```
 
 The boot snapshot is context for session start. For any subsequent status question, the DB is the source of truth. Query it. `--brief-only` is fast (read-only, no reconciliation, no side effects).