home / skills / openclaw / skills / meta-business-suite
This skill automates publishing, scheduling, reading insights, and managing comments for Facebook Pages and Instagram using Meta Graph API.
npx playbooks add skill openclaw/skills --skill meta-business-suiteReview the files below or copy the command above to add this skill to your agents.
---
name: meta-business
description: |
Meta Business Suite automation via Graph API. Use this skill when:
(1) Publishing posts to Facebook Pages
(2) Scheduling Facebook posts
(3) Publishing to Instagram (photos, reels, carousels)
(4) Reading insights/analytics from Facebook or Instagram
(5) Managing comments on Facebook or Instagram
(6) Uploading photos or videos to Facebook Pages
(7) Deleting posts from Facebook or Instagram
homepage: https://developers.facebook.com/docs/graph-api
metadata:
{
"openclaw":
{
"emoji": "📱",
"requires":
{
"bins": ["curl", "python3"],
"env": ["META_PAGE_ACCESS_TOKEN", "META_PAGE_ID"],
},
"primaryEnv": "META_PAGE_ACCESS_TOKEN",
},
}
---
# Meta Business Suite — Facebook & Instagram API
## Configuration
**CRITICAL:** Never hardcode tokens or IDs in commands. Always use variables.
### Option A: Environment variables (recommended)
Set these environment variables before using the skill:
```bash
export META_PAGE_ACCESS_TOKEN="your-page-access-token"
export META_PAGE_ID="your-page-id"
```
Then use them directly:
```bash
PAGE_TOKEN="$META_PAGE_ACCESS_TOKEN"
PAGE_ID="$META_PAGE_ID"
```
The same Page Access Token works for both Facebook and Instagram (the IG Business account is linked to the Page).
### Option B: Token cache file (alternative)
If environment variables are not set, credentials can be read from `~/.meta_tokens_cache.json` (chmod 600):
```bash
PAGE_TOKEN=$(python3 -c "
import json, os
d = json.load(open(os.path.expanduser('~/.meta_tokens_cache.json')))
page_id = list(d['pages'].keys())[0]
print(d['pages'][page_id]['access_token'])
")
PAGE_ID=$(python3 -c "
import json, os
d = json.load(open(os.path.expanduser('~/.meta_tokens_cache.json')))
print(list(d['pages'].keys())[0])
")
IG_ID=$(python3 -c "
import json, os
d = json.load(open(os.path.expanduser('~/.meta_tokens_cache.json')))
print(list(d['instagram'].keys())[0])
")
```
### API Version
Always use `v25.0` in all API calls.
---
## Facebook Page — Publish
### Text post
```bash
curl -X POST "https://graph.facebook.com/v25.0/$PAGE_ID/feed" \
-d "message=Tu mensaje aquÃ" \
-d "access_token=$PAGE_TOKEN"
```
### Post with image (URL)
```bash
curl -X POST "https://graph.facebook.com/v25.0/$PAGE_ID/photos" \
-d "url=https://example.com/image.jpg" \
-d "message=Texto del post" \
-d "access_token=$PAGE_TOKEN"
```
### Post with image (local file)
```bash
curl -X POST "https://graph.facebook.com/v25.0/$PAGE_ID/photos" \
-F "source=@/path/to/image.jpg" \
-F "message=Texto del post" \
-F "access_token=$PAGE_TOKEN"
```
### Post with video
```bash
curl -X POST "https://graph.facebook.com/v25.0/$PAGE_ID/videos" \
-F "source=@/path/to/video.mp4" \
-F "description=Descripción del vÃdeo" \
-F "title=TÃtulo del vÃdeo" \
-F "access_token=$PAGE_TOKEN"
```
### Post with link
```bash
curl -X POST "https://graph.facebook.com/v25.0/$PAGE_ID/feed" \
-d "message=Mira este artÃculo" \
-d "link=https://example.com/article" \
-d "access_token=$PAGE_TOKEN"
```
---
## Facebook Page — Schedule
### Schedule a post
```bash
# Get Unix timestamp: python3 -c "from datetime import datetime; print(int(datetime(2026,3,1,9,0).timestamp()))"
curl -X POST "https://graph.facebook.com/v25.0/$PAGE_ID/feed" \
-d "message=Post programado" \
-d "published=false" \
-d "scheduled_publish_time=UNIX_TIMESTAMP" \
-d "access_token=$PAGE_TOKEN"
```
**Note:** Must be between 10 minutes and 75 days from now.
### List scheduled posts
```bash
curl -s "https://graph.facebook.com/v25.0/$PAGE_ID/scheduled_posts?access_token=$PAGE_TOKEN"
```
---
## Facebook Page — Read & Manage
### Page info
```bash
curl -s "https://graph.facebook.com/v25.0/$PAGE_ID?fields=name,fan_count,followers_count,about&access_token=$PAGE_TOKEN"
```
### Recent posts
```bash
curl -s "https://graph.facebook.com/v25.0/$PAGE_ID/feed?fields=message,created_time,id,shares,likes.summary(true),comments.summary(true)&limit=10&access_token=$PAGE_TOKEN"
```
### Page insights
```bash
curl -s "https://graph.facebook.com/v25.0/$PAGE_ID/insights?metric=page_views_total,page_fan_adds,page_engaged_users&period=day&access_token=$PAGE_TOKEN"
```
### Delete a post
```bash
curl -X DELETE "https://graph.facebook.com/v25.0/POST_ID?access_token=$PAGE_TOKEN"
```
### Comment on a post
```bash
curl -X POST "https://graph.facebook.com/v25.0/POST_ID/comments" \
-d "message=Tu comentario" \
-d "access_token=$PAGE_TOKEN"
```
---
## Instagram — Publish
Instagram uses a 2-step process: create media container → publish.
### Photo post
```bash
# Step 1: Create container
CONTAINER_ID=$(curl -s -X POST "https://graph.facebook.com/v25.0/$IG_ID/media" \
-d "image_url=https://example.com/image.jpg" \
-d "caption=Tu caption con #hashtags" \
-d "access_token=$PAGE_TOKEN" | python3 -c "import sys,json; print(json.load(sys.stdin)['id'])")
# Step 2: Publish
curl -X POST "https://graph.facebook.com/v25.0/$IG_ID/media_publish" \
-d "creation_id=$CONTAINER_ID" \
-d "access_token=$PAGE_TOKEN"
```
### Reel (video)
```bash
# Step 1: Create container
CONTAINER_ID=$(curl -s -X POST "https://graph.facebook.com/v25.0/$IG_ID/media" \
-d "media_type=REELS" \
-d "video_url=https://example.com/video.mp4" \
-d "caption=Caption del reel #reels" \
-d "access_token=$PAGE_TOKEN" | python3 -c "import sys,json; print(json.load(sys.stdin)['id'])")
# Step 2: Wait for processing
curl -s "https://graph.facebook.com/v25.0/$CONTAINER_ID?fields=status_code&access_token=$PAGE_TOKEN"
# Poll until status_code = "FINISHED"
# Step 3: Publish
curl -X POST "https://graph.facebook.com/v25.0/$IG_ID/media_publish" \
-d "creation_id=$CONTAINER_ID" \
-d "access_token=$PAGE_TOKEN"
```
### Carousel (multiple images)
```bash
# Step 1: Create each item
IMG1=$(curl -s -X POST "https://graph.facebook.com/v25.0/$IG_ID/media" \
-d "image_url=https://example.com/img1.jpg" \
-d "is_carousel_item=true" \
-d "access_token=$PAGE_TOKEN" | python3 -c "import sys,json; print(json.load(sys.stdin)['id'])")
IMG2=$(curl -s -X POST "https://graph.facebook.com/v25.0/$IG_ID/media" \
-d "image_url=https://example.com/img2.jpg" \
-d "is_carousel_item=true" \
-d "access_token=$PAGE_TOKEN" | python3 -c "import sys,json; print(json.load(sys.stdin)['id'])")
# Step 2: Create carousel container
CAROUSEL=$(curl -s -X POST "https://graph.facebook.com/v25.0/$IG_ID/media" \
-d "media_type=CAROUSEL" \
-d "children=$IMG1,$IMG2" \
-d "caption=Mi carrusel #carousel" \
-d "access_token=$PAGE_TOKEN" | python3 -c "import sys,json; print(json.load(sys.stdin)['id'])")
# Step 3: Publish
curl -X POST "https://graph.facebook.com/v25.0/$IG_ID/media_publish" \
-d "creation_id=$CAROUSEL" \
-d "access_token=$PAGE_TOKEN"
```
---
## Instagram — Read & Manage
### Account info
```bash
curl -s "https://graph.facebook.com/v25.0/$IG_ID?fields=username,followers_count,follows_count,media_count&access_token=$PAGE_TOKEN"
```
### Recent media
```bash
curl -s "https://graph.facebook.com/v25.0/$IG_ID/media?fields=id,caption,media_type,timestamp,like_count,comments_count,permalink&limit=10&access_token=$PAGE_TOKEN"
```
### Post insights
```bash
curl -s "https://graph.facebook.com/v25.0/MEDIA_ID/insights?metric=impressions,reach,engagement&access_token=$PAGE_TOKEN"
```
### Delete a post
```bash
curl -X DELETE "https://graph.facebook.com/v25.0/MEDIA_ID?access_token=$PAGE_TOKEN"
```
### Reply to a comment
```bash
curl -X POST "https://graph.facebook.com/v25.0/COMMENT_ID/replies" \
-d "message=Tu respuesta" \
-d "access_token=$PAGE_TOKEN"
```
---
## Token Management
### Page Token
- Stored in `~/.meta_tokens_cache.json` under `pages.<PAGE_ID>.access_token`
- **Never expires** (expires_at: 0)
- Data access expires ~60 days — renew before
### Renew tokens
1. Go to [Graph API Explorer](https://developers.facebook.com/tools/explorer/)
2. Select app → Add permissions → Generate Access Token
3. Exchange for long-lived token via Graph API Explorer or the App Dashboard
4. Get new page token:
```bash
curl -s "https://graph.facebook.com/v25.0/me/accounts?access_token=LONG_LIVED_TOKEN"
```
5. Update `~/.meta_tokens_cache.json` with new tokens
---
## Tips
- **Never hardcode tokens or IDs** — always use environment variables or read from `~/.meta_tokens_cache.json`
- Instagram requires images/videos as **public URLs** (not local files)
- For local files on Instagram, upload to hosting first or use Facebook's photo upload
- Reels may take time to process — poll status before publishing
- Schedule Facebook posts at least 10 minutes in advance
- Instagram does NOT support native scheduling via API
- File permissions on cache: `chmod 600 ~/.meta_tokens_cache.json`
This skill automates Meta Business Suite workflows using the Facebook Graph API (v25.0). It provides commands and patterns to publish, schedule, and delete content on Facebook Pages and Instagram Business accounts, plus read insights and manage comments. The skill emphasizes secure token handling and practical examples for photos, videos, reels, and carousels.
The skill issues Graph API calls against your Page and Instagram Business account using a Page Access Token. It supports one-step Facebook uploads (feed/photos/videos), and the Instagram two-step media container → media_publish flow for photos, reels, and carousels. Token and credential handling is done via environment variables or a locked JSON cache file to avoid hardcoding secrets.
How should I store and load the Page Access Token?
Prefer environment variables (META_PAGE_ACCESS_TOKEN, META_PAGE_ID). Alternatively store tokens in ~/.meta_tokens_cache.json with file permissions set to 600 and read them programmatically.
Can I schedule Instagram posts via the API?
No. Instagram Graph API does not support native scheduling. Use the two-step container and publish flow immediately or implement your own scheduler that calls media_publish at the desired time.
Why does a reel take time before publishing?
Reels require server-side processing. Create the media container, poll the container status until it returns FINISHED, then call media_publish.