The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to perform diagnostic REST API queries against a ServiceNow instance.
Evidence: Steps 1 through 5 in SKILL.md provide curl command examples for querying tables such as svc_graph_connector, sys_import_set_run, and cmdb_ci.
Scope: The commands are limited to standard, authenticated ServiceNow API endpoints for data retrieval.
[DYNAMIC_EXECUTION]: The skill provides server-side JavaScript templates intended for execution via the SN-Execute-Background-Script tool to perform health checks and CI remediation.
Evidence: Step 6 and Step 7 in SKILL.md include script blocks for generating health reports and merging duplicate Configuration Items.
Execution method: These scripts are passed as parameters to the ServiceNow MCP tool for server-side processing on the instance.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from ServiceNow tables that may contain logs or error messages originating from external integration sources.
Ingestion points: SKILL.md (reading records from svc_graph_connector.error_message, discovery_log.message, and sys_import_set_row.sys_row_error).
Boundary markers: The skill does not define specific boundary markers or instructions to ignore embedded content within the log data.
Capability inventory: The skill has the capability to write records via SN-Update-Record and execute scripts via SN-Execute-Background-Script.
Sanitization: No explicit sanitization or filtering of the processed log content is described.

service-graph-diagnosis