The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local PowerShell script (cfe-borrow.ps1) using powershell.exe via the Bash tool. It passes parameters provided by the user (-ExtensionPath, -ConfigPath, -Object). If the underlying script does not properly sanitize these inputs, it could be vulnerable to command injection.
[DYNAMIC_EXECUTION]: The skill generates new source code files (Module.bsl) and metadata files (Form.xml) by extracting and transforming content from a source configuration. While this is a core feature, it constitutes automated dynamic creation of executable artifacts.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external configuration files to determine its logic. 1. Ingestion points: Reads .v8-project.json from the project root and 1C configuration files (e.g., Configuration.xml) from the path specified in -ConfigPath. 2. Boundary markers: The instructions do not define specific delimiters or warnings to ignore embedded instructions within these data sources. 3. Capability inventory: The skill has the ability to read files, perform file globbing, and execute PowerShell scripts via the Bash tool. 4. Sanitization: No sanitization or validation of the input configuration data is specified, which could allow maliciously crafted project files to influence agent actions.

cfe-borrow