The Agent Skills Directory

[COMMAND_EXECUTION]: The skill automates the development lifecycle by executing dotnet commands for project scaffolding, compilation, and unit testing. It also uses PowerShell scripts to orchestrate integration tests within the AutoCAD accoreconsole environment.
[DATA_EXFILTRATION]: The skill interacts with official Autodesk APIs at developer.api.autodesk.com to manage AppBundles and Activities. These operations are essential for the migration workflow. The skill follows security best practices by requiring users to store sensitive API credentials in local, gitignored .env files rather than requesting them in the conversation.
[INDIRECT_PROMPT_INJECTION]: By processing untrusted .lsp files to generate C# code, the skill has an indirect prompt injection surface. Malicious content within the LISP files could attempt to influence the generated output or the agent's reasoning.
Ingestion points: AutoLISP source files processed in Step 1.
Boundary markers: The skill uses a structured 'Discovery Table' to map LISP patterns to .NET equivalents.
Capability inventory: Access to dotnet and PowerShell tools, network operations to Autodesk APIs, and file system read/write.
Sanitization: Relies on strict pattern mapping as defined in the provided reference materials.
[DYNAMIC_EXECUTION]: The skill generates C# source code and project configuration files at runtime, which are subsequently compiled into executable assemblies.
[EXTERNAL_DOWNLOADS]: Standard development dependencies are resolved from NuGet package registries during the build phase.

lisp-to-dotnet