The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external ServiceNow tables such as alm_hardware and alm_license. This creates an attack surface where malicious instructions embedded in asset records could influence the agent's behavior.
Ingestion points: The skill utilizes snow_query_table and snow_cmdb_search to retrieve records from the database (found in SKILL.md).
Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat retrieved data as untrusted or to ignore instructions within that data.
Capability inventory: The skill leverages tools like snow_create_asset and provides examples for snow_execute_script (found in SKILL.md), allowing the agent to perform write operations and execute server-side scripts.
Sanitization: The provided instructions do not include specific steps for sanitizing, escaping, or validating the content retrieved from external tables before processing.
[DYNAMIC_EXECUTION]: The skill documents the use of snow_execute_script, which allows for the execution of arbitrary server-side JavaScript (GlideRecord scripts) within the ServiceNow environment. While this is a standard administrative capability, it involves generating and executing code based on templates and agent-determined logic.

asset-management