The Agent Skills Directory

[COMMAND_EXECUTION]: The skill interacts with Salesforce organizations using the sf CLI. The script scripts/upsert-and-report.sh executes standard commands such as sf data query, sf org display, and sf api request rest. These operations are limited to the user's authenticated environment and represent the intended functionality of the skill.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external data through positional arguments and environment variables (e.g., developer-name, overflow-assignee). To mitigate injection risks, the upsert-and-report.sh script performs strict regex validation on these inputs before interpolating them into SOQL queries or REST API request bodies. For instance, the DEVELOPER_NAME must match ^[A-Za-z][A-Za-z0-9_]{0,79}$, and OVERFLOW_ASSIGNEE_INPUT is validated against Salesforce ID and Username patterns.
[DYNAMIC_EXECUTION]: The test suite includes scripts/tests/_bootstrap.py, which uses subprocess.run to execute the skill's bash scripts within a controlled environment. This is a standard software development practice for unit testing and does not involve executing code from untrusted or remote sources.
[SAFE]: The skill implements a production environment guard. It queries the target organization's metadata to verify if it is a Sandbox, Trial, or Developer Edition before proceeding with any write operations, blocking execution on standard production instances to prevent accidental data modification.

service-omni-queue-routing-config-deploy