The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from several external ServiceNow tables (proc_po, proc_po_item, core_company, etc.) which acts as untrusted input.
Ingestion points: Data is retrieved using tools like SN-Query-Table and SN-Get-Record as described in SKILL.md.
Boundary markers: There are no explicit instructions or delimiters provided to the agent to distinguish between its core instructions and the external data being processed.
Capability inventory: The skill has access to the Bash shell and various ServiceNow API tools.
Sanitization: No data sanitization or escaping techniques are mentioned for handling the retrieved content before it is processed by the model.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute curl commands for interacting with ServiceNow REST API endpoints. This is a standard procedure for the skill's intended procurement tasks.

purchase-order-summarization