The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the sf CLI and a specific Node.js script located in the agent's environment.
Evidence: Multiple instances of sf data360 query ... and node ~/.claude/skills/sf-datacloud/scripts/diagnose-org.mjs are used throughout SKILL.md and README.md to interact with the target environment.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources (Salesforce Data Cloud SQL results, metadata, and search index listings) which could potentially contain malicious instructions intended to influence the agent's behavior.
Ingestion points: SQL query results from sf data360 query sql, table descriptions from sf data360 query describe, and search index lists from sf data360 search-index list in SKILL.md.
Boundary markers: None specified in the instructions to delimit or ignore instructions within the retrieved data.
Capability inventory: The skill has the capability to execute shell commands (sf, node) and perform further introspection based on the data it reads.
Sanitization: There is no evidence of sanitization or filtering of the content retrieved from the Data Cloud before it is presented to or processed by the agent.

sf-datacloud-retrieve