import os, requests
YT = os.environ["YOUTUBE_API_KEY"]
MV = os.environ["MAVERA_API_KEY"]
YT_BASE = "https://www.googleapis.com/youtube/v3"
MV_BASE = "https://app.mavera.io/api/v1"
MV_H = {"Authorization": f"Bearer {MV}", "Content-Type": "application/json"}
BRAND_CONTEXT = """
Brand: [Your Brand] — DTC fitness/wellness company
Channel focus: Workout tutorials, nutrition science, product reviews
Audience: 18-35, health-conscious, 60% female, US-based
Current content gaps we suspect: trending challenges, reaction content, collaborations
"""
# 1. Fetch top 50 trending videos (1 quota unit per call)
trending = requests.get(f"{YT_BASE}/videos", params={
"key": YT, "chart": "mostPopular",
"part": "snippet,contentDetails,statistics,topicDetails",
"maxResults": 50, "regionCode": "US",
}).json()
if "error" in trending:
raise SystemExit(f"YouTube API error: {trending['error']['message']}")
videos = []
category_counts = {}
all_tags = []
for item in trending.get("items", []):
snippet = item["snippet"]
stats = item.get("statistics", {})
category = snippet.get("categoryId", "0")
tags = snippet.get("tags", [])
category_counts[category] = category_counts.get(category, 0) + 1
all_tags.extend(tags[:10])
videos.append({
"title": snippet["title"],
"channel": snippet["channelTitle"],
"category": category,
"tags": tags[:10],
"views": int(stats.get("viewCount", 0)),
"likes": int(stats.get("likeCount", 0)),
"description": snippet.get("description", "")[:200],
})
# 2. Resolve category IDs to names
cat_ids = ",".join(set(category_counts.keys()))
categories = requests.get(f"{YT_BASE}/videoCategories", params={
"key": YT, "id": cat_ids, "part": "snippet",
}).json()
cat_map = {
item["id"]: item["snippet"]["title"]
for item in categories.get("items", [])
}
# 3. Build trending summary
trending_block = "\n".join(
f"- \"{v['title']}\" by {v['channel']} | Category: {cat_map.get(v['category'], v['category'])} | "
f"Views: {v['views']:,} | Tags: {', '.join(v['tags'][:5])}"
for v in sorted(videos, key=lambda x: -x["views"])[:30]
)
category_summary = "\n".join(
f"- {cat_map.get(cid, cid)}: {count} videos"
for cid, count in sorted(category_counts.items(), key=lambda x: -x[1])
)
# 4. Gap analysis via Mave
gap = requests.post(f"{MV_BASE}/mave/chat", headers=MV_H, json={
"message": f"""Identify content gaps between what's trending on YouTube and what our brand covers.
BRAND CONTEXT:
{BRAND_CONTEXT}
TRENDING NOW (top 30 by views):
{trending_block}
CATEGORY DISTRIBUTION:
{category_summary}
TOP TRENDING TAGS: {', '.join(set(all_tags)[:30])}
Produce:
1. **Trending themes** our brand is NOT covering (ranked by opportunity size)
2. **Specific video concepts** for each gap (title, format, estimated audience size)
3. **Cross-category opportunities** — trends in non-fitness categories our fitness brand could adapt
4. **Timing urgency** — which gaps are time-sensitive (trend peaking) vs. evergreen
5. **Risk assessment** — which trends could backfire for our brand""",
}).json()
print("TRENDING CONTENT GAP ANALYSIS")
print("=" * 60)
print(f"Analyzed: {len(videos)} trending videos across {len(category_counts)} categories")
print(f"Top categories: {', '.join(cat_map.get(c, c) for c, _ in sorted(category_counts.items(), key=lambda x: -x[1])[:5])}")
print("\n" + gap.get("content", "")[:2500])
