The Agent Skills Directory

[SAFE]: The skill's primary purpose is performing inventory and dependency analysis of OmniStudio components (OmniScripts, FlexCards, Integration Procedures, and Data Mappers).
[COMMAND_EXECUTION]: Uses the official Salesforce CLI (sf) to perform metadata queries. The commands are standard SOQL probes (sf data query) used to detect which Salesforce namespace (Core, vlocity_cmt, or vlocity_ins) is active in the target org.
[DATA_EXPOSURE_&_EXFILTRATION]: No evidence of data exfiltration or unauthorized access. All queries target OmniStudio configuration metadata objects. The skill does not access sensitive local file paths like .ssh, .aws, or .env files.
[PROMPT_INJECTION]: The instructions do not contain markers intended to bypass safety guidelines, override agent behavior, or extract system prompts. Instructional language like "CRITICAL: Orchestration Order" refers to the sequence of operations for technical analysis rather than an attempt to override AI constraints.
[REMOTE_CODE_EXECUTION]: No external code is downloaded or executed. The skill relies entirely on pre-installed CLI tools and built-in analytical logic.
[INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface (parsing OmniStudio JSON metadata fields like PropertySetConfig). However, the analysis is limited to extracting specific keys for dependency mapping. There are no high-privilege write operations or shell execution paths that consume this external data.

omnistudio-dependencies-analyze