The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The documentation explicitly mentions and encourages the use of a curl -fsSL ... | bash pattern for installation from https://aihot.virxact.com/aihot-skill/install.sh. This pattern is a significant security risk as it executes unverified remote scripts directly in the user's shell environment.
[COMMAND_EXECUTION]: The skill provides complex shell commands for the agent to run, including date calculations for time windows and curl with specific headers and jq filtering. This indicates the skill is designed to trigger command execution within the agent's environment to process data.
[EXTERNAL_DOWNLOADS]: The skill is designed to fetch data from an external, non-whitelisted domain (aihot.virxact.com) via API calls to retrieve news items.
[INDIRECT_PROMPT_INJECTION]: The skill retrieves untrusted content (titles, summaries, and news items) from a remote API and presents it to the agent for summarization and formatting without clear boundary protection.
Ingestion points: API responses from https://aihot.virxact.com/api/public/* containing user-controlled news data.
Boundary markers: None identified. The agent is not instructed to use delimiters or specific safety instructions to ignore potential commands embedded in the news content.
Capability inventory: Uses curl, jq, and date via shell execution.
Sanitization: There is no evidence of sanitizing or escaping the API response before the agent processes the natural language content for output.

aihot