The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes instructions for the agent to execute shell scripts (update-skill.sh or update-skill.ps1) and local shell probes to verify installation paths and perform version checks.\n- [EXTERNAL_DOWNLOADS]: The update scripts fetch configuration and code updates from the official vendor repository at github.com/gate/gate-skills using standard tools like curl, wget, or git.\n- [REMOTE_CODE_EXECUTION]: The skill features a self-update mechanism that can download and replace local skill files with newer versions from the remote source. This process is gated by explicit instructions requiring the agent to obtain user confirmation and verify a security token before applying changes.\n- [INDIRECT_PROMPT_INJECTION]: The skill aggregates market news, social sentiment, and event data from external MCP tools, which presents a surface for processing untrusted content.\n
Ingestion points: Data enters the context via news_events_get_latest_events, news_feed_search_news, and news_feed_get_social_sentiment in SKILL.md.\n
Boundary markers: The agent is constrained by a specific 'Report Template' and 'Safety Rules' that require source attribution and de-duplication.\n
Capability inventory: The skill possesses file system write capabilities (via update scripts) and network access across all provided scripts.\n
Sanitization: Logic for de-duplication, impact-based ranking, and objective summarization is mandated to prevent adversarial content from overriding intended behavior.

gate-news-briefing