The Agent Skills Directory

EXTERNAL_DOWNLOADS (MEDIUM): The skill utilizes npx @salesforce/b2c-cli, which downloads and executes a package from the npm registry at runtime. Although the package belongs to a known organization (Salesforce), it is not within the explicitly predefined trusted list, making it an unverifiable external dependency.- COMMAND_EXECUTION (SAFE): The skill is designed to execute standard b2c CLI commands for job management. This aligns with its primary stated purpose and does not appear to involve arbitrary or malicious command construction.- DATA_EXFILTRATION (LOW): The job export and job import commands facilitate the movement of site archives and data between the local environment and the Salesforce cloud. While this is intended behavior, it represents a data movement vector that should be monitored.- PROMPT_INJECTION (LOW): The skill has a surface for Indirect Prompt Injection (Category 8). An attacker who can influence job names or log output on the Salesforce instance could potentially inject instructions that the agent might follow when processing the results of b2c job search or b2c job run --show-log.
Ingestion points: b2c job search output and job log contents via --show-log.
Boundary markers: Absent; the agent is not instructed to ignore instructions within the logs.
Capability inventory: CLI execution and file system access (job import).
Sanitization: None; the skill does not specify how to sanitize or escape data retrieved from the instance logs.

b2c-job