The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates interaction with external ITSM records, which creates a surface for indirect prompt injection if an attacker embeds malicious instructions in incident fields that the agent later processes.
Ingestion points: Data enters the agent context via tools like SN-List-Incidents, SN-Get-Incident, and SN-Natural-Language-Search (as defined in the tools metadata and referenced in SKILL.md).
Boundary markers: The instructions do not provide delimiters or "ignore embedded instructions" warnings for data retrieved from ServiceNow records.
Capability inventory: The skill has significant write capabilities across the ServiceNow environment, including SN-Update-Record, SN-Create-Incident, SN-Assign-Incident, and SN-Close-Incident (as defined in the tools metadata and referenced in SKILL.md).
Sanitization: No sanitization, escaping, or validation logic is specified for handling external content before interpolation or processing.

quick-reference