The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts (scripts/study_sources.py) and the scrapling CLI tool to perform environment checks and data extraction tasks.
[EXTERNAL_DOWNLOADS]: The documentation suggests initializing the firecrawl-cli using npx, which involves downloading and executing code from the NPM registry.
[DATA_EXFILTRATION]: The study_sources.py script performs network requests to api.tikhub.io and api.tikhub.dev to retrieve research data. These operations are restricted to public routes and do not involve the transmission of sensitive local files.
[PROMPT_INJECTION]: The skill acts as an ingestion layer for untrusted external data, creating an attack surface for indirect prompt injection.
Ingestion points: Social media content and web pages are retrieved via TikHub API, Firecrawl, and Scrapling.
Boundary markers: The skill mandates separating raw evidence but lacks explicit boundary delimiters or 'ignore' instructions for the agent when processing the research.v1 packets.
Capability inventory: The skill maintains network access capabilities and the ability to execute CLI tools.
Sanitization: While the script normalizes numeric metrics and uses standard JSON parsing, it does not include specific sanitization or filtering for the textual content of the scraped evidence.

study-scraping