The Agent Skills Directory

[SAFE]: The skill does not contain any detected malicious patterns, obfuscation, or unauthorized network operations. Its behavior is consistent with its stated purpose as a developer productivity tool.
[COMMAND_EXECUTION]: The skill uses grep -rl to identify target content.json files and ls to verify the presence of components in a local temporary directory. These operations are limited to project file discovery and do not involve executing arbitrary or untrusted code.
[PROMPT_INJECTION]: The skill processes content.json files from the local environment, which represents an indirect prompt injection surface. However, the risk is minimized by the use of structured JSON parsing and specific instructions that limit the agent's modification to the "definition" field only.
Ingestion points: The skill reads content.json files located in the sfdc_cms__view/ and sfdc_cms__themeLayout/ directories of the Salesforce project.
Boundary markers: Instructions explicitly constrain the agent to preserve all JSON properties and only modify "definition" values using an authoritative mapping.
Capability inventory: Access to Bash (grep, ls), Read, and Write tools for local file management.
Sanitization: The skill relies on structured JSON tool interactions rather than raw text manipulation, reducing the likelihood of processing malicious payloads as instructions.

commerce-b2b-open-code-components-replace