luke/ai-podcast
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Publishing, website, reaper, and doc updates

Browse Source 
Publish script, clip maker, website worker + data, reaper lua helpers,
audio settings, and CLAUDE.md doc reorg.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
luke
2026-04-09 23:39:57 -06:00
co-authored by Claude Opus 4.6
parent 44ac569fec
commit 61b3cba412
10 changed files with 398 additions and 127 deletions
Download Patch File Download Diff File Expand all files Collapse all files
CLAUDE.md
+38 -34
View File
@@ -1,20 +1,5 @@
# AI Podcast - Project Instructions
## Git Remote (Gitea)
- **Repo**: `git@gitea-nas:luke/ai-podcast.git`
- **Web**: http://mmgnas:3000/luke/ai-podcast
- **SSH Host**: `gitea-nas` (configured in ~/.ssh/config)
- HostName: `mmgnas` (use `mmgnas-10g` if wired connection issues)
- Port: `2222`
- User: `git`
- IdentityFile: `~/.ssh/gitea_mmgnas`
## NAS Access
- **Hostname**: `mmgnas` (wireless) or `mmgnas-10g` (wired/10G)
- **SSH Port**: 8001
- **User**: luke
- **Docker path**: `/share/CACHEDEV1_DATA/.qpkg/container-station/bin/docker`
## Castopod (Podcast Publishing)
- **URL**: https://podcast.macneilmediagroup.com
- **Podcast handle**: `@LukeAtTheRoost`
@@ -42,8 +27,7 @@ Required in `.env`:
- ELEVENLABS_API_KEY (optional)
- INWORLD_API_KEY (for Inworld TTS)
## Post-Production Pipeline (added Feb 2026)
- **Branch**: `feature/real-callers` — all current work is here, pushed to gitea
## Post-Production Pipeline
- **Stem Recorder** (`backend/services/stem_recorder.py`): Records 5 WAV stems (host, caller, music, sfx, ads) during live shows. Uses lock-free deque architecture — audio callbacks just append to deques, a background writer thread drains to disk. `write()` for continuous streams (host mic, music, ads), `write_sporadic()` for burst sources (caller TTS, SFX) with time-aligned silence padding.
- **Audio hooks** in `backend/services/audio.py`: 7 tap points guarded by `if self.stem_recorder:`. Persistent mic stream (`start_stem_mic`/`stop_stem_mic`) runs during recording to capture host voice continuously, not just during push-to-talk.
- **API endpoints**: `POST /api/recording/start`, `POST /api/recording/stop` (auto-runs postprod in background thread), `POST /api/recording/process`
@@ -88,24 +72,44 @@ Required in `.env`:
- **Analytics**: Cloudflare Web Analytics (enable in Cloudflare dashboard, no code changes needed)
- **Deploy**: `npx wrangler pages deploy website/ --project-name=lukeattheroost --branch=main`
## Git Push
- If `mmgnas` times out, use the 10g hostname:
```bash
GIT_SSH_COMMAND="ssh -o HostName=mmgnas-10g -p 2222 -i ~/.ssh/gitea_mmgnas" git push origin main
```
## Hetzner VPS
- **IP**: `46.225.164.41`
- **SSH**: `ssh root@46.225.164.41` (uses default key `~/.ssh/id_rsa`)
- **Specs**: 2 CPU, 4GB RAM, 38GB disk (~33GB free)
- **Mail**: `docker-mailserver` at `/opt/mailserver/`
- **Manage accounts**: `docker exec mailserver setup email add/del/list`
- **Available for future services** — has headroom for lightweight containers. Not suitable for storage-heavy services (e.g. Castopod with daily episodes) without a disk upgrade or attached volume.
## Podcast Workflow
- Publishing pipeline: episodes go through Castopod, CDN, website, YouTube, and social
- Always check Python venv is active and packages are installed before running publish scripts
- Episode numbering must be verified against existing episodes
- Episode numbering: check Castopod for the latest episode number, don't hardcode
## Episodes Published
- Episode 6 published 2026-02-08 (podcast6.mp3, ~31 min)
## Scripts
- `publish_episode.py` — Transcribes audio, generates metadata (title, description, cover art), publishes to Castopod. Usage: `python publish_episode.py ~/Desktop/episode.mp3`
- `make_clips.py` — Two-pass clip extraction: fast Whisper transcription → LLM selects best moments → quality Whisper re-transcription for precise timestamps. Usage: `python make_clips.py ~/Desktop/episode.mp3 --count 3`
- `generate_milestone_images.py` — Generates social milestone images via Gemini Flash (requires GOOGLE_API_KEY)
- `post_milestone.py` — Posts milestone announcements to social platforms via Postiz
- `make_x_launch_assets.py` — Generates branded visual assets for X/Twitter (header, quote cards, intro/review graphics)
- `schedule_x_launch.py` — Schedules X/Twitter launch campaign posts via Postiz API
## Reaper Scripts
- `reaper/dialog_regions.lua` — Background script that polls `/tmp/reaper_state.txt` and creates colored regions (green=DIALOG, red=AD, blue=IDENT) as the backend writes state changes during recording
- `reaper/strip_silence_dialog.lua` — Post-production script: strips long silences from dialog regions, normalizes AD/IDENT/music volume, trims music to voice length with fade-out, mutes music during AD/IDENT regions
## Cost Dashboard
- **Route**: `/costs` — standalone analytics page, linked from control panel header
- **Database**: `data/costs.db` (SQLite) — aggregates all session cost data for cross-session queries
- **Data layer**: `backend/services/cost_db.py` — schema, JSON import, all query functions
- **Dual-write**: `cost_tracker.py` writes to both JSON (`data/cost_reports/`) and SQLite on every LLM/TTS call
- **API**: 8 endpoints under `/api/costs/` — summary, timeline, models, categories, sessions, session detail, expensive calls, TTS providers
- **Frontend**: `frontend/costs.html`, `frontend/css/costs.css`, `frontend/js/costs.js` — Chart.js for visualizations
- **Pricing**: Hardcoded in `cost_tracker.py` (`OPENROUTER_PRICING`, `TTS_PRICING`) — update when provider prices change
- **Not tracked yet**: SignalWire call costs
## Data Directory
State files (not config — these are written at runtime):
- `regulars.json` — Returning caller profiles (backgrounds, key moments, arc status, relationships)
- `used_topics_history.json` — Previously used caller topics to avoid repeats
- `session_checkpoint.json` — Current show session state (call history, caller queue)
- `publish_state.json` — Publishing pipeline progress per episode
- `intern.json` — Devon's lookup history
- `emails.json` — Listener email submissions
- `voicemails.json` — Listener voicemail submissions
## Personal
- Don't build anything until you have 95% clarity on what I want you to do. Ask clarifying questions until you reach 95% understanding of what I'm asking
- When working as a team, propose the plan before executing — don't just start building
- Flag trade-offs that affect show quality or listener experience rather than silently resolving them
audio_settings.json
+3 -2
View File
@@ -1,10 +1,11 @@
{
"input_device": 13,
"input_device": 14,
"input_device_name": "Babyface Pro (70793771)",
"input_channel": 1,
"output_device": 12,
"output_device": 13,
"output_device_name": "Radio Voice Mic",
"caller_channel": 3,
"devon_channel": 17,
"live_caller_channel": 9,
"music_channel": 5,
"sfx_channel": 7,
make_clips.py
+2 -2
View File
@@ -1249,10 +1249,10 @@ def generate_clip_video_remotion(
]
try:
result = subprocess.run(cmd, capture_output=True, text=True, cwd=str(REMOTION_DIR), timeout=180)
result = subprocess.run(cmd, capture_output=True, text=True, cwd=str(REMOTION_DIR), timeout=600)
except subprocess.TimeoutExpired:
props_path.unlink(missing_ok=True)
print(f" Remotion render timed out (180s)")
print(f" Remotion render timed out (600s)")
return False
props_path.unlink(missing_ok=True)
publish_episode.py
+146 -81
View File
@@ -204,12 +204,12 @@ TRANSCRIPT:
"Content-Type": "application/json"
},
json={
"model": "anthropic/claude-3.5-sonnet",
"model": "anthropic/claude-sonnet-4.6",
"messages": [{"role": "user", "content": full_prompt}],
"max_tokens": 8192,
"temperature": 0
},
timeout=120
timeout=300
)
except requests.exceptions.Timeout:
print(f" Warning: Speaker labeling timed out for chunk {i+1}, using raw text")
@@ -1311,19 +1311,23 @@ def upload_to_youtube(audio_path: str, metadata: dict, chapters: list,
video_path = Path(audio_path).with_suffix(".yt.mp4")
# Convert MP3 + cover art to MP4 (pad to 1920x1080 for YouTube compatibility)
print(" Converting audio to video...")
result = subprocess.run([
"ffmpeg", "-y", "-loop", "1",
"-i", str(cover_art), "-i", audio_path,
"-vf", "scale=-1:1080,pad=1920:1080:(ow-iw)/2:0:black",
"-c:v", "libx264", "-tune", "stillimage",
"-c:a", "aac", "-b:a", "192k",
"-pix_fmt", "yuv420p", "-shortest",
"-movflags", "+faststart", str(video_path)
], capture_output=True, text=True, timeout=1800)
if result.returncode != 0:
print(f" Warning: ffmpeg failed: {result.stderr[-200:]}")
return None
# Skip if video already exists and is non-trivial size (>1MB)
if video_path.exists() and video_path.stat().st_size > 1_000_000:
print(f" Using existing video: {video_path} ({video_path.stat().st_size / 1_000_000:.0f}MB)")
else:
print(" Converting audio to video...")
result = subprocess.run([
"ffmpeg", "-y", "-loop", "1",
"-i", str(cover_art), "-i", audio_path,
"-vf", "scale=-1:1080,pad=1920:1080:(ow-iw)/2:0:black",
"-c:v", "libx264", "-tune", "stillimage",
"-c:a", "aac", "-b:a", "192k",
"-pix_fmt", "yuv420p", "-shortest",
"-movflags", "+faststart", str(video_path)
], capture_output=True, text=True, timeout=1800)
if result.returncode != 0:
print(f" Warning: ffmpeg failed: {result.stderr[-200:]}")
return None
# Build chapter timestamps for description
chapter_lines = []
@@ -1446,6 +1450,7 @@ def main():
parser.add_argument("--title", "-t", help="Override generated title")
parser.add_argument("--description", help="Override generated description")
parser.add_argument("--session-data", "-s", help="Path to session export JSON (from /api/session/export)")
parser.add_argument("--resume", action="store_true", help="Resume a failed publish — skip transcription/Castopod, continue from CDN/YouTube/social")
args = parser.parse_args()
audio_path = Path(args.audio_file).expanduser().resolve()
@@ -1493,88 +1498,148 @@ def main():
episode_number = get_next_episode_number()
print(f"Episode number: {episode_number}")
# Guard against duplicate publish
if not args.dry_run:
exists = _check_episode_exists_in_db(episode_number)
if exists is None:
print(f"Error: Could not reach Castopod DB to check for duplicates. "
f"Aborting to prevent duplicate uploads. Fix NAS connectivity and retry.")
_cleanup_mysql_auth()
lock_fp.close()
LOCK_FILE.unlink(missing_ok=True)
sys.exit(1)
if exists:
print(f"Error: Episode {episode_number} already exists in Castopod. "
f"Use --episode-number to specify a different number, or remove the existing episode first.")
# --- Resume path: skip transcription + Castopod, pick up from CDN/YouTube/social ---
if args.resume:
castopod_step = _get_step_details(episode_number, "castopod")
if not castopod_step:
print(f"Error: No Castopod data in publish_state.json for episode {episode_number}. Nothing to resume.")
_cleanup_mysql_auth()
lock_fp.close()
LOCK_FILE.unlink(missing_ok=True)
sys.exit(1)
episode = {"id": castopod_step["episode_id"], "slug": castopod_step["slug"]}
print(f"Resuming episode {episode_number}: id={episode['id']}, slug={episode['slug']}")
# Load session data if provided
session_data = None
if args.session_data:
session_path = Path(args.session_data).expanduser().resolve()
if session_path.exists():
with open(session_path) as f:
session_data = json.load(f)
print(f"Loaded session data: {session_data.get('call_count', 0)} calls")
# Load existing metadata from files saved during first run
transcript_path = audio_path.with_suffix(".transcript.txt")
chapters_path = audio_path.with_suffix(".chapters.json")
srt_path = audio_path.with_suffix(".srt")
# Build metadata from existing files; re-transcribe + gen only if missing
if chapters_path.exists():
print(" Loading existing chapters and metadata files")
with open(chapters_path) as f:
chapters_data = json.load(f)
# Reconstruct metadata from slug and chapters
slug = episode["slug"]
# Convert slug to title: "episode-49-foo-bar" -> "Episode 49: Foo Bar"
title_part = slug.replace(f"episode-{episode_number}-", "").replace("-", " ").title()
metadata = {
"title": f"Episode {episode_number}: {title_part}",
"description": f"Episode {episode_number} of Luke at the Roost.",
"chapters": chapters_data if isinstance(chapters_data, list) else chapters_data.get("chapters", []),
"thumbnail_text": title_part.upper()[:30],
}
else:
print(f"Warning: Session data file not found: {session_path}")
print("[1/5] Re-transcribing audio for metadata...")
transcript = transcribe_audio(str(audio_path))
metadata = generate_metadata(transcript, episode_number)
save_chapters(metadata, str(chapters_path))
if not transcript_path.exists():
labeled_text = label_transcript_speakers(transcript["full_text"])
with open(transcript_path, "w") as f:
f.write(labeled_text)
if not srt_path.exists():
generate_srt(transcript["segments"], str(srt_path))
# Step 1: Transcribe
transcript = transcribe_audio(str(audio_path))
if args.title:
metadata["title"] = args.title
if args.description:
metadata["description"] = args.description
# Step 2: Generate metadata
metadata = generate_metadata(transcript, episode_number)
srt_path = audio_path.with_suffix(".srt")
direct_upload = os.path.getsize(str(audio_path)) > CLOUDFLARE_UPLOAD_LIMIT
chapters_uploaded = True
transcript_uploaded = True
yt_video_id = None
# Jump to CDN upload (step 3.7)
# (fall through to the common CDN/publish/YouTube/social code below)
# Use session chapters if available (more accurate than LLM-generated)
if session_data and session_data.get("chapters"):
metadata["chapters"] = session_data["chapters"]
print(f" Using {len(metadata['chapters'])} chapters from session data")
else:
# --- Normal (non-resume) path ---
# Apply overrides
if args.title:
metadata["title"] = args.title
if args.description:
metadata["description"] = args.description
# Guard against duplicate publish
if not args.dry_run:
exists = _check_episode_exists_in_db(episode_number)
if exists is None:
print(f"Error: Could not reach Castopod DB to check for duplicates. "
f"Aborting to prevent duplicate uploads. Fix NAS connectivity and retry.")
_cleanup_mysql_auth()
lock_fp.close()
LOCK_FILE.unlink(missing_ok=True)
sys.exit(1)
if exists:
print(f"Error: Episode {episode_number} already exists in Castopod. "
f"Use --episode-number to specify a different number, or remove the existing episode first.")
_cleanup_mysql_auth()
lock_fp.close()
LOCK_FILE.unlink(missing_ok=True)
sys.exit(1)
# Save chapters file
chapters_path = audio_path.with_suffix(".chapters.json")
save_chapters(metadata, str(chapters_path))
# Load session data if provided
session_data = None
if args.session_data:
session_path = Path(args.session_data).expanduser().resolve()
if session_path.exists():
with open(session_path) as f:
session_data = json.load(f)
print(f"Loaded session data: {session_data.get('call_count', 0)} calls")
else:
print(f"Warning: Session data file not found: {session_path}")
# Save transcript text file with LUKE:/CALLER: speaker labels
transcript_path = audio_path.with_suffix(".transcript.txt")
raw_text = transcript["full_text"]
labeled_text = label_transcript_speakers(raw_text)
with open(transcript_path, "w") as f:
f.write(labeled_text)
print(f" Transcript saved to: {transcript_path}")
# Step 1: Transcribe
transcript = transcribe_audio(str(audio_path))
# Generate SRT from whisper segments (for Castopod/podcast apps)
srt_path = audio_path.with_suffix(".srt")
generate_srt(transcript["segments"], str(srt_path))
print(f" SRT saved to: {srt_path}")
# Step 2: Generate metadata
metadata = generate_metadata(transcript, episode_number)
# Save session transcript alongside episode if available (has speaker labels)
if session_data and session_data.get("transcript"):
session_transcript_path = audio_path.with_suffix(".session_transcript.txt")
with open(session_transcript_path, "w") as f:
f.write(session_data["transcript"])
print(f" Session transcript saved to: {session_transcript_path}")
# Use session chapters if available (more accurate than LLM-generated)
if session_data and session_data.get("chapters"):
metadata["chapters"] = session_data["chapters"]
print(f" Using {len(metadata['chapters'])} chapters from session data")
if args.dry_run:
print("\n[DRY RUN] Would publish with:")
print(f" Title: {metadata['title']}")
print(f" Description: {metadata['description']}")
print(f" Chapters: {json.dumps(metadata['chapters'], indent=2)}")
print("\nChapters file saved. Run without --dry-run to publish.")
return
# Apply overrides
if args.title:
metadata["title"] = args.title
if args.description:
metadata["description"] = args.description
# Step 3: Create episode
direct_upload = os.path.getsize(str(audio_path)) > CLOUDFLARE_UPLOAD_LIMIT
episode = create_episode(str(audio_path), metadata, episode_number, duration=transcript["duration"])
_mark_step_done(episode_number, "castopod", {"episode_id": episode["id"], "slug": episode.get("slug")})
# Save chapters file
chapters_path = audio_path.with_suffix(".chapters.json")
save_chapters(metadata, str(chapters_path))
# Save transcript text file with LUKE:/CALLER: speaker labels
transcript_path = audio_path.with_suffix(".transcript.txt")
raw_text = transcript["full_text"]
labeled_text = label_transcript_speakers(raw_text)
with open(transcript_path, "w") as f:
f.write(labeled_text)
print(f" Transcript saved to: {transcript_path}")
# Generate SRT from whisper segments (for Castopod/podcast apps)
srt_path = audio_path.with_suffix(".srt")
generate_srt(transcript["segments"], str(srt_path))
print(f" SRT saved to: {srt_path}")
# Save session transcript alongside episode if available (has speaker labels)
if session_data and session_data.get("transcript"):
session_transcript_path = audio_path.with_suffix(".session_transcript.txt")
with open(session_transcript_path, "w") as f:
f.write(session_data["transcript"])
print(f" Session transcript saved to: {session_transcript_path}")
if args.dry_run:
print("\n[DRY RUN] Would publish with:")
print(f" Title: {metadata['title']}")
print(f" Description: {metadata['description']}")
print(f" Chapters: {json.dumps(metadata['chapters'], indent=2)}")
print("\nChapters file saved. Run without --dry-run to publish.")
return
# Step 3: Create episode
direct_upload = os.path.getsize(str(audio_path)) > CLOUDFLARE_UPLOAD_LIMIT
episode = create_episode(str(audio_path), metadata, episode_number, duration=transcript["duration"])
_mark_step_done(episode_number, "castopod", {"episode_id": episode["id"], "slug": episode.get("slug")})
# Step 3.5: Upload chapters and transcript to Castopod
# (must happen before CDN sync so media records exist for syncing)
reaper/dialog_regions.lua
+2 -1
View File
@@ -24,7 +24,8 @@ local last_state = ""
local transport_active = false
local function log(msg)
reaper.ShowConsoleMsg("[Regions] " .. msg .. "\n")
-- Silent by default — uncomment for debugging:
-- reaper.ShowConsoleMsg("[Regions] " .. msg .. "\n")
end
local function is_playing_or_recording()
reaper/strip_silence_dialog.lua
+2 -1
View File
@@ -30,7 +30,8 @@ local BLOCK_SAMPLES = math.floor(SAMPLE_RATE * BLOCK_SEC)
local THRESHOLD = 10 ^ (SILENCE_DB / 20)
local MIN_VOICE_BLOCKS = math.ceil(MIN_VOICE_SEC / BLOCK_SEC)
local function log(msg)
reaper.ShowConsoleMsg("[PostProd] " .. msg .. "\n")
-- Silent by default — uncomment for debugging:
-- reaper.ShowConsoleMsg("[PostProd] " .. msg .. "\n")
end
---------------------------------------------------------------------------
website/_worker.js
+20 -3
View File
@@ -58,9 +58,26 @@ export default {
return new Response("Feed unavailable", { status: 502 });
}
// Plausible analytics proxy (bypass ad blockers)
// Umami analytics proxy (bypass ad blockers)
if (url.pathname === "/api/send" && request.method === "POST") {
const body = await request.text();
const resp = await fetch("https://plausible.macneilmediagroup.com/api/send", {
method: "POST",
headers: {
"Content-Type": "application/json",
"User-Agent": request.headers.get("User-Agent") || "",
"X-Forwarded-For": request.headers.get("CF-Connecting-IP") || request.headers.get("X-Forwarded-For") || "",
},
body,
});
return new Response(resp.body, {
status: resp.status,
headers: { "Content-Type": resp.headers.get("Content-Type") || "text/plain" },
});
}
if (url.pathname === "/p/script") {
const resp = await fetch("https://plausible.macneilmediagroup.com/js/script.file-downloads.hash.outbound-links.pageview-props.revenue.tagged-events.js");
const resp = await fetch("https://plausible.macneilmediagroup.com/script.js");
return new Response(await resp.text(), {
headers: {
"Content-Type": "application/javascript",
@@ -71,7 +88,7 @@ export default {
if (url.pathname === "/p/event" && request.method === "POST") {
const body = await request.text();
const resp = await fetch("https://plausible.macneilmediagroup.com/api/event", {
const resp = await fetch("https://plausible.macneilmediagroup.com/api/send", {
method: "POST",
headers: {
"Content-Type": "application/json",
website/data/clips.json
+118 -1
View File
@@ -1,4 +1,121 @@
[
{
"title": "Thinking in English First",
"description": "When she heard about her father's diagnosis, she processed it in English first\u2014before responding to her mom in Czech. It felt like a betrayal of who she used to be.",
"episode_number": 49,
"clip_file": "clip-5-thinking-in-english-first.mp4",
"youtube_id": "LGqpEAxe754",
"featured": false,
"thumbnail": "images/clips/clip-5-thinking-in-english-first.jpg"
},
{
"title": "The April Fool's Divorce Text Disaster",
"description": "She thought sending her husband a divorce text as an April Fool's joke would be funny. Then his mom got involved and the casseroles started arriving. For 11 days straight.",
"episode_number": 48,
"clip_file": "clip-2-the-april-fool-s-divorce-text-disaster.mp4",
"youtube_id": "L9XAQgZeQo0",
"featured": false,
"thumbnail": "images/clips/clip-2-the-april-fool-s-divorce-text-disaster.jpg"
},
{
"title": "Caller's Daughter on Adult Website Revealed",
"description": "A caller discovers his daughter's secret online activities in the worst possible way. The details just keep getting more uncomfortable.",
"episode_number": 47,
"clip_file": "clip-3-caller-s-daughter-on-adult-website-revealed.mp4",
"youtube_id": "LiH6gVDTHcg",
"featured": false,
"thumbnail": "images/clips/clip-3-caller-s-daughter-on-adult-website-revealed.jpg"
},
{
"title": "The Taint Explanation",
"description": "Luke's caller had to explain what a taint is on live radio and it went exactly as uncomfortably as you'd imagine. This is why we can't have nice things.",
"episode_number": 46,
"clip_file": "clip-2-the-taint-explanation.mp4",
"youtube_id": "7qAhmnwUE1c",
"featured": false,
"thumbnail": "images/clips/clip-2-the-taint-explanation.jpg"
},
{
"title": "Bear Traps and Pit of Snakes",
"description": "Luke's home security system is straight out of Home Alone but way more dangerous. Swinging logs? Check. Pit of snakes covered with dust? BOOM! \ud83d\udc0d",
"episode_number": 44,
"clip_file": "clip-2-bear-traps-and-pit-of-snakes.mp4",
"youtube_id": "XBSEu2bFUAI",
"featured": false,
"thumbnail": "images/clips/clip-2-bear-traps-and-pit-of-snakes.jpg"
},
{
"title": "Roommate's Age Play Kink Takes Over Kitchen",
"description": "Roommate's age play kink is now front and center in the shared kitchen. Baby bottles by the coffee maker? Pacifiers on the counter? This living situation just got complicated.",
"episode_number": 43,
"clip_file": "clip-3-roommate-s-age-play-kink-takes-over-kitchen.mp4",
"youtube_id": "5cF_O2Gm9yM",
"featured": false,
"thumbnail": "images/clips/clip-3-roommate-s-age-play-kink-takes-over-kitchen.jpg"
},
{
"title": "Day Trading Away the Marriage",
"description": "She's been pretending to commute for 12 hours a day while secretly day-trading their life savings away in motels. Two years of lies and $145K gone.",
"episode_number": 42,
"clip_file": "clip-1-day-trading-away-the-marriage.mp4",
"youtube_id": "JeZ9OLGfW8A",
"featured": false,
"thumbnail": "images/clips/clip-1-day-trading-away-the-marriage.jpg"
},
{
"title": "The Snot Boogie Beer Money Thief",
"description": "He showed up to the barbecue claiming he's 'not like other guys' then literally ran off with everyone's beer money. The audacity is unmatched.",
"episode_number": 42,
"clip_file": "clip-2-the-snot-boogie-beer-money-thief.mp4",
"youtube_id": "taBAo8_YMkA",
"featured": false,
"thumbnail": "images/clips/clip-2-the-snot-boogie-beer-money-thief.jpg"
},
{
"title": "Aliens Chasing in Ford Trucks",
"description": "Glowing Ford trucks with shadowy drivers that aren't quite human. This caller's alien encounter story is straight out of a fever dream.",
"episode_number": 41,
"clip_file": "clip-3-aliens-chasing-in-ford-trucks.mp4",
"youtube_id": "Io8CzDtKGfA",
"featured": false,
"thumbnail": "images/clips/clip-3-aliens-chasing-in-ford-trucks.jpg"
},
{
"title": "Doctor Prescribes Prostate Stimulation",
"description": "Dale's doctor hands him a prostate stimulation pamphlet after his cancer diagnosis. Luke's medical advice? 'Stick some stuff up there and see what works.'",
"episode_number": 40,
"clip_file": "clip-3-doctor-prescribes-prostate-stimulation.mp4",
"youtube_id": "nQtwJpocFpY",
"featured": false,
"thumbnail": "images/clips/clip-3-doctor-prescribes-prostate-stimulation.jpg"
},
{
"title": "Open Marriage Backfired Spectacularly",
"description": "He agreed to an open marriage and now his wife is living her best life while he sits home alone every night. This call is PAINFUL to listen to.",
"episode_number": 39,
"clip_file": "clip-1-open-marriage-backfired-spectacularly.mp4",
"youtube_id": "-K-t7iijfGs",
"featured": false,
"thumbnail": "images/clips/clip-1-open-marriage-backfired-spectacularly.jpg"
},
{
"title": "Cat Burglar Bringing Home Stolen Goods",
"description": "This caller's cat has been stealing from the neighbors and bringing home cash and car keys. Where is this criminal mastermind getting this stuff?",
"episode_number": 39,
"clip_file": "clip-2-cat-burglar-bringing-home-stolen-goods.mp4",
"youtube_id": "JvgJWxFCBZk",
"featured": false,
"thumbnail": "images/clips/clip-2-cat-burglar-bringing-home-stolen-goods.jpg"
},
{
"title": "Second Baby Shower Entitlement Rant",
"description": "A second baby shower with a full registry? This caller is NOT having it and goes OFF about the entitlement of expecting gifts for baby number two.",
"episode_number": 39,
"clip_file": "clip-3-second-baby-shower-entitlement-rant.mp4",
"youtube_id": "NKt8NjDHKcg",
"featured": false,
"thumbnail": "images/clips/clip-3-second-baby-shower-entitlement-rant.jpg"
},
{
"title": "Cult Leader Realizes He's Been Manipulating People",
"description": "Cult leader calls in having a full existential crisis about his 'shared intimacy nights' and the manipulation tactics he's been using on his followers.",
@@ -55,7 +172,7 @@
},
{
"title": "Started a Fight and Can't Stop Reading About Wars",
"description": "A caller starts a fight with their partner and spirals into an obsessive deep-dive on historical wars. Luke tries to untangle the connection.",
"description": "",
"episode_number": 31,
"clip_file": "clip-3-started-a-fight-and-can-t-stop-reading-about-wars.mp4",
"youtube_id": "D2iWnSGQeow",
website/index.html
+1 -2
View File
@@ -92,8 +92,7 @@
}
}]
</script>
<script defer data-domain="lukeattheroost.com" data-api="/p/event" src="/p/script"></script>
<script>window.plausible = window.plausible || function() { (window.plausible.q = window.plausible.q || []).push(arguments) }</script>
<script defer data-website-id="ae816554-19c8-4e97-a2cb-d0226c27e9bc" data-host-url="/" src="/p/script"></script>
</head>
<body>
website/sitemap.xml
+66
View File
@@ -270,4 +270,70 @@
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
<url>
<loc>https://lukeattheroost.com/episode.html?slug=episode-39-st-patrick-s-day-chaos-and-caller-confessions</loc>
<lastmod>2026-03-18</lastmod>
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
<url>
<loc>https://lukeattheroost.com/episode.html?slug=episode-40-prostate-cancer-christmas-lights-and-potato-salad-betrayals</loc>
<lastmod>2026-03-19</lastmod>
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
<url>
<loc>https://lukeattheroost.com/episode.html?slug=episode-41-benny-s-creepy-vhs-tape-from-nowhere</loc>
<lastmod>2026-03-20</lastmod>
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
<url>
<loc>https://lukeattheroost.com/episode.html?slug=episode-42-peggy-s-day-trading-disaster-and-the-nonprofit-yacht</loc>
<lastmod>2026-03-21</lastmod>
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
<url>
<loc>https://lukeattheroost.com/episode.html?slug=episode-43-cousin-rufus-and-the-locksmith-s-wild-weekend</loc>
<lastmod>2026-03-23</lastmod>
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
<url>
<loc>https://lukeattheroost.com/episode.html?slug=episode-44-floyd-s-fence-and-the-cattle-conspiracy</loc>
<lastmod>2026-03-24</lastmod>
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
<url>
<loc>https://lukeattheroost.com/episode.html?slug=episode-45-potato-salad-aliens-and-the-dripping-faucet-of-doom</loc>
<lastmod>2026-03-26</lastmod>
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
<url>
<loc>https://lukeattheroost.com/episode.html?slug=episode-46-butchers-wreak-forever-and-other-late-night-confessions</loc>
<lastmod>2026-03-30</lastmod>
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
<url>
<loc>https://lukeattheroost.com/episode.html?slug=episode-47-clarence-randy-and-the-onlyfans-daughter</loc>
<lastmod>2026-03-30</lastmod>
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
<url>
<loc>https://lukeattheroost.com/episode.html?slug=episode-48-potato-salad-legacy-and-the-april-fool-s-apocalypse</loc>
<lastmod>2026-04-01</lastmod>
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
<url>
<loc>https://lukeattheroost.com/episode.html?slug=episode-49-silas-s-shared-intimacy-night-and-four-brave-souls</loc>
<lastmod>2026-04-05</lastmod>
<changefreq>never</changefreq>
<priority>0.7</priority>
</url>
</urlset>