The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill mandates that the agent verify volatile facts such as model versions, prices, and laws against live external sources. This ingestion of untrusted data from the web or official docs exposes the agent to indirect prompt injection.
Ingestion points: External fact verification and documentation review as specified in SKILL.md.
Boundary markers: Absent; the instructions do not provide delimiters or warnings to ignore instructions embedded in fetched content.
Capability inventory: The skill assumes access to shell tools (e.g., date) and web browsing capabilities for verification.
Sanitization: Absent; no instructions are provided for sanitizing or validating external content before processing.
[COMMAND_EXECUTION]: The skill contains specific templates for executing shell commands to retrieve system time information.
Evidence: SKILL.md provides the example TZ=Pacific/Auckland date.

temporal-awareness