The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill performs network requests to api.scrapecreators.com to search for TikTok videos and fetch Instagram reels. This behavior is the primary function of the skill and is clearly documented in SKILL.md and the reference files.
[DATA_EXFILTRATION]: The skill manages authentication using the SCRAPE_CREATORS_API_KEY environment variable, which is a standard and secure practice for secret management. The scripts do not access any sensitive local file paths (e.g., SSH keys, AWS credentials) and only transmit data to the designated API endpoint.
[PROMPT_INJECTION]: The skill processes untrusted content from social media platforms, creating a potential surface for indirect prompt injection.
Ingestion points: Untrusted social media data (captions, hashtags, user descriptions) enters the agent's context through the JSON and summary outputs of scripts/get_instagram_reels.py and scripts/search_tiktok_keyword.py.
Boundary markers: The output is structured as JSON or formatted tables, but does not include explicit boundary markers or instructions to the agent to ignore embedded commands.
Capability inventory: The skill's capabilities are restricted to network requests to the Scrape Creators API and printing data to standard output. It does not have access to file-write operations, system configuration, or arbitrary command execution.
Sanitization: The scripts perform basic text formatting, such as stripping newlines and truncating descriptions in summary modes, though no formal instruction-sanitization logic is present.

social-media-scrape