The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the sail CLI to execute environment management and API commands such as sail api get and sail api post. These operations are essential for retrieving the data required for security analysis.
[PROMPT_INJECTION]: The skill processes identity attributes and role metadata from the SailPoint environment, creating a surface for indirect prompt injection. (1) Ingestion points: Data fetched from SailPoint endpoints like /v2025/sod-policies and /v2025/search. (2) Boundary markers: There are no instructions to use delimiters or specific safety wrappers when processing tenant data. (3) Capability inventory: The skill is restricted to using the sail CLI for API calls and does not perform broad file-system or network operations. (4) Sanitization: No data sanitization is performed on the results returned from the SailPoint API before analysis.

sailpoint-toxic-access-detector