中文 EN

Scan Report

Scan New Files

⚠ Low Risk — Risk Score 18/100
Last scan:23 hr ago Rescan
18 /100
instagram-posting-pipeline
Generate promotional images and post them to Instagram Business accounts via Meta Graph API. Handles Pillow image generation, Cloudinary hosting, and Meta Graph API two-step posting.
This is a legitimate Instagram marketing automation pipeline with no malicious behavior. All network, filesystem, and credential access are declared and necessary for its documented purpose.
Skill Nameinstagram-posting-pipeline
Duration37.3s
Enginepi
✓
Safe to install
Skill is safe to use. Replace hardcoded placeholder credential strings in SKILL.md documentation with generic placeholders to avoid confusion. Consider pinning Pillow version in any future dependency file.

Findings 2 items

Severity Finding Location
Low
Placeholder credential examples in documentation Doc Mismatch
SKILL.md lines 26 and 128 contain example ACCESS_TOKEN strings ('your_page_access_token' and 'token_for_account_a') in the environment setup documentation. While these are clearly example placeholders (labeled as such), they could cause confusion and may trigger security scanners. The values are not actual secrets.
export IG_ACCESS_TOKEN="your_page_access_token"
→ Replace with '<YOUR_PAGE_ACCESS_TOKEN>' or similar to clearly indicate a placeholder that must be replaced.
 SKILL.md:26
Low
Web scraper extracts email and phone via regex Sensitive Access
scrape_business.py uses regex to extract email addresses and phone numbers from scraped websites. This is within scope for a business info scraper, but the extracted contact data is returned in plaintext. No exfiltration observed.
email_match = re.search(r"[\w.+-]+@[\w-]+\.[\w.-]+", html_content)
→ No action needed — this is legitimate functionality for the declared use case. Consider adding a note that extracted contact data is not transmitted anywhere.
 scripts/scrape_business.py:101
ResourceDeclaredInferredStatusEvidence
Filesystem WRITE WRITE ✓ Aligned generate_course_promo.py:39 — img.save() to configured OUTPUT_DIR
Network READ READ ✓ Aligned post_to_instagram.py, upload_cloudinary.py, scrape_business.py — all HTTPS API c…
Shell NONE NONE — No subprocess or shell invocation found
Environment READ READ ✓ Aligned os.environ.get() used to read IG_ACCESS_TOKEN, IG_BUSINESS_ACCOUNT_ID, CLOUDINAR…
2 High 9 findings
🔑
High API Key 疑似硬编码凭证
ACCESS_TOKEN="your_page_access_token"
SKILL.md:26
🔑
High API Key 疑似硬编码凭证
ACCESS_TOKEN="token_for_account_a"
SKILL.md:128
🔗
Medium External URL 外部 URL
https://developers.facebook.com/tools/explorer/
SKILL.md:42
🔗
Medium External URL 外部 URL
https://res.cloudinary.com/demo/image/upload/xyz.png
SKILL.md:85
🔗
Medium External URL 外部 URL
https://graph.facebook.com/v21.0/
scripts/post_to_instagram.py:39
🔗
Medium External URL 外部 URL
https://www.instagram.com/p/
scripts/post_to_instagram.py:71
🔗
Medium External URL 外部 URL
https://www.caddeskcentre.com
scripts/scrape_business.py:168
🔗
Medium External URL 外部 URL
https://cloudinary.com
scripts/upload_cloudinary.py:40
🔗
Medium External URL 外部 URL
https://api.cloudinary.com/v1_1/
scripts/upload_cloudinary.py:57

File Tree

5 files · 23.3 KB · 682 lines
Python 4f · 551L Markdown 1f · 131L
├─ ▾ 📁 scripts
│ ├─ 🐍 generate_course_promo.py Python 180L · 6.0 KB
│ ├─ 🐍 post_to_instagram.py Python 100L · 3.7 KB
│ ├─ 🐍 scrape_business.py Python 173L · 6.1 KB
│ └─ 🐍 upload_cloudinary.py Python 98L · 3.5 KB
└─ 📝 SKILL.md Markdown 131L · 4.0 KB

Dependencies 1 items

PackageVersionSourceKnown VulnsNotes
Pillow unpinned import No Pillow is not pinned in any requirements file. The script uses 'from PIL import Image'. Ensure Pillow is installed in the target environment.

Security Positives

✓ No subprocess or shell execution found — all logic uses Python stdlib
✓ No base64 obfuscation or dynamic code execution (no eval, exec, ast.literal_eval misuse)
✓ SSRF protection implemented in scrape_business.py — blocks localhost, private IPs, and internal network hostnames
✓ Cloudinary upload requires explicit env vars (no accidental demo cloud uploads) — security-conscious default
✓ All network access is declared in SKILL.md and is necessary for the pipeline's stated purpose
✓ No access to sensitive paths (~/.ssh, ~/.aws, .env)
✓ No credential exfiltration — IG_ACCESS_TOKEN is used only for Meta Graph API authentication
✓ No persistence mechanisms (no cron, startup hooks, or backdoor installation)
✓ No supply chain risk — only depends on Pillow (standard, trusted package)
✓ No hidden HTML comments or embedded instructions
✓ No direct IP network requests — all outbound calls go to well-known domains (graph.facebook.com, api.cloudinary.com)