The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute bundled local Python validation scripts (e.g., validate_flow_cli.py, simulate_flow.py) to perform 110-point static analysis and governor limit simulations on Salesforce Flow metadata. These operations are performed using the local Python interpreter as part of the skill's core functionality.
[INDIRECT_PROMPT_INJECTION]: The skill processes Salesforce Flow metadata retrieved at runtime from external orgs, which represents a vulnerability surface for indirect prompt injection.
Ingestion points: Flow XML and JSON metadata are ingested via the metadata_read, metadata_list, and tooling_api_query tools, as well as through user-supplied arguments (SKILL.md).
Boundary markers: Absent; the instructions do not define delimiters or specific "ignore embedded instructions" warnings for the external Flow content being analyzed.
Capability inventory: The skill possesses high-privilege capabilities to create, modify, and delete metadata and records in the connected Salesforce environment using tools such as metadata_create, metadata_update, tooling_api_dml, and sobject_dml (SKILL.md).
Sanitization: Absent; the skill does not explicitly describe sanitization or filtering of instructions that may be embedded in the fetched XML/JSON metadata before it enters the agent's context.

sf-flow