Telegram PDF Scraper
Analysis
Review recommended: the skill largely matches its Telegram PDF scraping purpose, but it stores a Telegram Web session locally and auto-clicks/downloads Telegram links while overstating its PDF-only safety.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
links = msg.locator("a").all() ... if not safe_filename.lower().endswith('.pdf'): safe_filename += ".pdf" ... link.click(force=True)The code scans all anchor links in messages, force-clicks them, and renames non-.pdf link text to .pdf rather than verifying the Telegram object type, file extension, MIME type, or user approval for each download.
**Anti-Malware:** This skill is hard-coded to ignore and block dangerous external links ... **Internal Only:** It only interacts with native Telegram "Document" objects ... download only `.pdf` files.
The safety wording is stronger than the implementation supports, because the code uses generic anchors and extension renaming rather than proving files are safe PDFs or native Telegram document objects.
playwright>=1.40.0
Playwright is purpose-aligned for Telegram Web automation, but the dependency is not pinned to an exact reviewed version.
Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.
user_data_dir = os.path.abspath("./openclaw_chrome_profile") ... p.chromium.launch_persistent_context(user_data_dir=user_data_dir, headless=False, accept_downloads=True)The skill creates a persistent browser profile that can hold the user's Telegram Web session, giving the automation delegated access to that Telegram account and retaining session state locally.
