The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes several untrusted data sources which could contain malicious instructions.
Ingestion points: Reads project-level files including packagedef, env.json, launch.json, and BSL source code files (.bsl) from the src/ directory.
Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands within these processed files.
Capability inventory: The skill has the ability to execute shell commands via oscript, write files to the project directory, and manage the IDE terminal.
Sanitization: There is no evidence of sanitization or validation of the content read from these external files before the agent interacts with them.
[COMMAND_EXECUTION]: The skill relies on the execution of oscript (OneScript) for building configurations, managing infobases, and disassembling files. The instructions provide examples of sending raw text to the terminal to execute these scripts.
[DYNAMIC_EXECUTION]: The skill facilitates the compilation and execution of external reports and processors (.erf/.epf) and includes a TypeScript API example that programmatically executes oscript build files at runtime.
[EXTERNAL_DOWNLOADS]: The documentation refers to external tools and repositories, including the OneScript engine download site and configuration schemas hosted on GitHub by the vanessa-opensource organization.

1c-devtools-cursor