The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute Salesforce CLI commands (sf) for environment verification, deployment promotion, and audit reporting. It also calls a local classification script (sf-deploy-gate) to programmatically verify that the target environment is a production instance before proceeding.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external organization metadata and local project files which could potentially contain malicious instructions if those sources were compromised.
Ingestion points: Untrusted data enters the context from the output of sf org display, sf project deploy report, and the contents of the .sfdx/last-validation.json file.
Boundary markers: The skill uses a formatted production confirmation banner to present the ingested data (Org alias, Instance, Edition, Job ID) to the user for manual verification.
Capability inventory: The skill possesses the capability to execute shell commands (Bash) and read local files, specifically performing metadata deployments to production orgs.
Sanitization: The skill implements a strict gate using a classification script to validate the organization type and explicitly requires the user to type 'yes' before executing the final deployment command.
[DATA_EXPOSURE]: The skill accesses the .sfdx directory to retrieve validation metadata and writes deployment reports to .sfdx/deploy-history/. These files contain project-specific metadata and job identifiers necessary for the deployment lifecycle.

platform-quick-deploy