The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external Salesforce debug logs, which constitutes an untrusted data source.
Ingestion points: Debug logs retrieved via sf apex log get or analyzed from .log files in the workspace.
Boundary markers: None identified. The instructions do not specify delimiters or instructions for the agent to ignore embedded commands within the log files.
Capability inventory: The skill has access to Bash(sf *) (Salesforce CLI), as well as Read, Write, and Edit file operations. It can execute Apex code via sf apex run and delete data via sf data delete bulk.
Sanitization: No explicit sanitization or validation of log content is described before processing.
[COMMAND_EXECUTION]: The skill frequently uses the sf (Salesforce CLI) for administrative tasks.
Evidence: Commands include sf apex log get, sf apex run, and sf data delete bulk. While these are restricted to the sf binary in allowed-tools, they provide significant control over the connected Salesforce environment.

sf-debug