The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill includes shell and PowerShell scripts (scripts/update-skill.sh and scripts/update-skill.ps1) designed to update the skill content.
The scripts download the latest version of the skill from the author's official GitHub repository (github.com/gate/gate-skills).
Update mechanisms include git clone, as well as downloading and expanding ZIP and tar.gz archives.
The scripts also perform a version check by fetching the SKILL.md frontmatter from raw.githubusercontent.com.
[COMMAND_EXECUTION]: The skill's primary function is to execute gate-cli (v0.5.2) commands to retrieve market intelligence.
Authorized commands include search-news, get-latest-events, get-exchange-announcements, search-ugc, and search-x.
[PROMPT_INJECTION]: The skill processes untrusted external data from news feeds and social media (UGC, X, Reddit, YouTube), which is a surface for indirect prompt injection.
Ingestion points: External data enters via the search-news, search-ugc, and search-x commands.
Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" markers for the untrusted data.
Capability inventory: The agent is restricted to executing gate-cli data-gathering tools.
Sanitization: The skill does not specify any sanitization or filtering of the retrieved content before it is presented to the agent for synthesis.

gate-news-intel