The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface as it ingests and processes untrusted content from external exchange announcement feeds.\n- Ingestion points: Data enters the context through the news_feed_get_exchange_announcements tool results referenced in SKILL.md.\n- Boundary markers: The skill lacks explicit boundary markers or instructions to the model to ignore embedded commands within the retrieved text.\n- Capability inventory: Maintenance scripts update-skill.sh and update-skill.ps1 provide network (curl, wget) and file-system (cp, rm, tar) capabilities, while the primary MCP tools are read-only.\n- Sanitization: No evidence of input validation or sanitization is present for the announcement data.\n- [EXTERNAL_DOWNLOADS]: The skill contains utility scripts (update-skill.sh and update-skill.ps1) that facilitate downloads and synchronization from the author's official GitHub repository (github.com/gate/gate-skills).\n- These scripts utilize git clone, unzip, and tar to maintain the skill's local files by fetching remote source code from the vendor infrastructure.

gate-news-listing