The Agent Skills Directory

[COMMAND_EXECUTION]: The main script scripts/upsert-and-report.sh performs operations using the Salesforce CLI (sf) and jq. User-provided inputs, such as Developer Names and User IDs, are filtered through strict regular expressions (e.g., ^[A-Za-z][A-Za-z0-9_]{0,79}$) before being used in shell commands or database queries.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data that could potentially contain malicious instructions.
Ingestion points: Positional arguments and environment variables (e.g., QRC_OVERFLOW_ASSIGNEE) passed to the upsert-and-report.sh script.
Boundary markers: None.
Capability inventory: Execution of sf CLI commands, which interact with the Salesforce API, and file system interactions via Bash tools.
Sanitization: Input parameters are validated against whitelist-based regular expressions to ensure they conform to expected formats (e.g., alphanumeric identifiers).
[DYNAMIC_EXECUTION]: The unit test infrastructure in scripts/tests/_bootstrap.py uses the subprocess module to run the skill's bash scripts. This is a controlled execution of local code for testing and does not process external, unsanitized strings as executable commands.

service-omni-queue-routing-config-deploy