Instagram SuperPower

Tools

Documentation and links

Path mapping

Agent machine

Cobalt server (self-hosted)

How it all connects

HikerAPI (cloud)                    Cobalt server (self-hosted)
─────────────────                    ────────────────────────────
api.instagrapi.com ◀── curl ──  Your machine
  analytics, data                    Cobalt API (localhost:9000)
  $0.001/request                       media download (free)

Your machine:
  analyze.sh  → HikerAPI (directly)
  download.sh → SSH → Cobalt server → Instagram CDN
  watchlist.sh → analyze.sh + download.sh

HikerAPI — Authentication

# IMPORTANT: use api.instagrapi.com (no Cloudflare) instead of api.hikerapi.com
# api.hikerapi.com blocks urllib/requests from scripts
cat ~/.secrets/hikerapi/api-key

# Usage in curl:
curl -s "https://api.instagrapi.com/ENDPOINT" \
  -H "accept: application/json" \
  -H "x-access-key: $(cat ~/.secrets/hikerapi/api-key)"

# Or as GET parameter:
curl -s "https://api.instagrapi.com/ENDPOINT?access_key=$(cat ~/.secrets/hikerapi/api-key)"

Balance: GET /sys/balance — shows remaining requests and balance.

HikerAPI — Main endpoints

Users

Media (posts/reels)

Search

Hashtags

Response data

Each media object contains:

• playcount / viewcount — views
• like_count — likes
• comment_count — comments
• taken_at — publish timestamp
• code — shortcode for URL (instagram.com/reel/{code}/)
• caption.text — caption text
• user.username — author

Quick start

Account profile

HIKER_KEY=$(cat ~/.secrets/hikerapi/api-key)
curl -s "https://api.instagrapi.com/v2/user/by/username?username=<username>" \
  -H "x-access-key: $HIKER_KEY" | python3 -m json.tool

Account reels (top performing)

HIKER_KEY=$(cat ~/.secrets/hikerapi/api-key)
USER_ID=$(curl -s "https://api.instagrapi.com/v2/user/by/username?username=<username>" \
  -H "x-access-key: $HIKER_KEY" | python3 -c "import json,sys; print(json.load(sys.stdin)['user']['pk'])")

curl -s "https://api.instagrapi.com/v2/user/clips?user_id=$USER_ID" \
  -H "x-access-key: $HIKER_KEY" | python3 -c "
import json, sys
d = json.load(sys.stdin)
items = d.get('response',{}).get('items',[])
for item in sorted(items, key=lambda x: x['media'].get('play_count',0), reverse=True)[:5]:
    m = item['media']
    print(f'{m.get(\"play_count\",0):>10,} views | {m.get(\"like_count\",0):>6,} likes | https://instagram.com/reel/{m[\"code\"]}/')
"

Download reel (via Cobalt)

bash ~/.claude/skills/instagram-superpower/scripts/download.sh "https://www.instagram.com/reel/<CODE>/"

Download + transcribe

bash ~/.claude/skills/instagram-superpower/scripts/download.sh "https://www.instagram.com/reel/<CODE>/" transcribe

Check HikerAPI balance

curl -s "https://api.instagrapi.com/sys/balance" \
  -H "x-access-key: $(cat ~/.secrets/hikerapi/api-key)"

Typical scenarios

1. "Show top reels of a competitor"

bash scripts/analyze.sh <username> 5 14

2. "Download this reel and summarize it"

bash scripts/download.sh "<url>" transcribe

Cobalt downloads → Groq transcribes → agent analyzes.

3. "Add to monitoring"

bash scripts/watchlist.sh add <username>
bash scripts/watchlist.sh scan 3 14

4. "Full competitor breakdown"

1. analyze.sh <username> 5 14 → top-5 reels + metrics
2. download.sh <url> transcribe for top-3
3. Agent analyzes: hook → pain → solution → CTA

5. "What's viral in a hashtag"

HIKER_KEY=$(cat ~/.secrets/hikerapi/api-key)
curl -s "https://api.instagrapi.com/v2/hashtag/by/name?name=aitools" -H "x-access-key: $HIKER_KEY"
curl -s "https://api.instagrapi.com/v2/hashtag/medias/top?hashtag_id=ID" -H "x-access-key: $HIKER_KEY"

Pagination

Many endpoints return nextpageid. For the next page:

curl -s "https://api.instagrapi.com/v2/user/clips?user_id=123&page_id=NEXT_PAGE_ID" \
  -H "x-access-key: $HIKER_KEY"

Reels: ~12 per page. For 50 reels you need ~4 requests.

Rate Limits

HikerAPI

• Limit: 8 requests/second (Standard plan)
• No risk of Instagram ban — HikerAPI uses its own account pool and proxies

Cobalt (downloads)

• Rate limits Instagram: 20 downloads/hour per 1 account
• Pause between requests: minimum 3 seconds
• Batch downloads (>5): pause 10-15 seconds
• Strictly sequential, never parallel

Optimal strategy

• Analytics (HikerAPI): no limits, up to 8 req/sec
• Downloads (Cobalt): 20/hour, sequential
• HikerAPI can also download (/v1/media/download/*), but uses paid requests — use Cobalt instead

Maintenance

HikerAPI

curl -s "https://api.instagrapi.com/sys/balance" -H "x-access-key: $(cat ~/.secrets/hikerapi/api-key)"
# Key in dashboard: https://hikerapi.com → Tokens

Cobalt

ssh root@<your-server-ip> "docker ps --filter name=cobalt-api --format '{{.Status}}'"
ssh root@<your-server-ip> "docker logs cobalt-api --tail 20"
ssh root@<your-server-ip> "cd /opt/cobalt && docker compose restart cobalt-api"

Updating Instagram cookies (for Cobalt)

1. Credentials: ~/.secrets/cobalt/accounts.json
2. Login via browser automation (agent-browser via CDP)
3. Login using Input.dispatchKeyEvent (character by character)
4. Button: div[role="button"]
5. 2FA: TOTP → input[name="verificationCode"]
6. Cookies: Network.getCookies → .instagram.com → cookies.json
7. Deploy: bash scripts/deploy-cookies.sh

Security

• HikerAPI key: ~/.secrets/hikerapi/api-key (chmod 600)
• Cobalt: ONLY 127.0.0.1:9000 on the Cobalt server
• Cookies/credentials: ONLY ~/.secrets/cobalt/ (chmod 600)
• Forbidden: logging keys/cookies, printing in chat, committing to git