The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local PowerShell and Python scripts to perform the decompilation task. This is the intended behavior and follows the standard operational model for such tools.
[PROMPT_INJECTION]: The skill performs indirect prompt injection as it processes untrusted XML data which is then presented to the agent for analysis and modification.
Ingestion points: The scripts in scripts/mxl-decompile.ps1 and scripts/mxl-decompile.py read an external XML file provided via the TemplatePath argument.
Boundary markers: Absent. The output is generated as a JSON DSL which may contain arbitrary text strings from the source XML without delimiters to distinguish between data and instructions.
Capability inventory: The skill uses Bash to execute the scripts and has access to Read and Write tools for file operations. The SKILL.md instructions explicitly tell the agent to analyze or modify the resulting JSON.
Sanitization: No sanitization or filtering is performed on the text content extracted from the XML items (parameters, templates, and text values).
[EXTERNAL_DOWNLOADS]: Both scripts contain a reference to the author's GitHub repository (github.com/Nikolay-Shirokov/cc-1c-skills) in the header comments. These are informative links and no code is downloaded at runtime.

mxl-decompile