The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted third-party data from Avito listings, which could contain malicious instructions designed to influence agent behavior.
Ingestion points: Untrusted data enters the context through avito_search and avito_card tools, which fetch listing descriptions and seller information from avito.ru (identified in SKILL.md).
Boundary markers: The instructions do not specify any delimiters or safety markers to isolate the external content from the agent's core instructions.
Capability inventory: The skill utilizes curl_cffi for network requests and interacts with Chrome over the Chrome DevTools Protocol (CDP).
Sanitization: There is no mention of sanitization or filtering applied to the external text before it is presented to the agent.
[EXTERNAL_DOWNLOADS]: The skill documentation references the use of curl_cffi, a Python library used to impersonate browser TLS signatures to bypass bot detection on the target site.

avito-connector