The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the bash tool to execute sf project generate manifest commands. It incorporates user-provided inputs such as source directories, metadata component names, and org aliases directly into these commands. This is standard behavior for the intended functionality.
[SAFE]: The skill interacts with local project files such as sfdx-project.json to resolve API versions and uses the Write tool to create XML files. All operations are consistent with Salesforce development workflows.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes user-supplied strings for manifest generation.
Ingestion points: User-provided paths for --source-dir and component names for the --metadata flag in SKILL.md.
Boundary markers: Absent; the instructions do not specify delimiters for user-provided data.
Capability inventory: The skill uses the bash tool for command execution and the Write tool for file creation.
Sanitization: No explicit sanitization or shell-escaping of user-provided strings is mandated in the instructions.

platform-manifest-generate