The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface as it ingests and processes data from external, potentially untrusted sources.\n
Ingestion points: The skill reads source code files (.cs), project definition files (.csproj, .sln, .slnx, Directory.Build.props, Directory.Packages.props), and configuration files (global.json, Dockerfile) from the local repository being migrated.\n
Boundary markers: The workflow includes an instruction to use loaded reference documents as authoritative, but lacks explicit delimiters or structural guards when interpolating content from the target repository into the agent's context.\n
Capability inventory: The skill utilizes significant capabilities including file system write operations, git commits, and execution of shell commands like dotnet build, dotnet restore, and dotnet test.\n
Sanitization: There is no evidence of sanitization or filtering of the ingested content before it is processed or executed during the build and test phases.\n- [COMMAND_EXECUTION]: The skill requires the execution of multiple shell commands to perform the migration and verification tasks.\n
Commands identified: dotnet build, dotnet restore, dotnet test, and dotnet --version are used throughout the workflow.\n
Context: These executions are directly tied to the primary purpose of the skill (migrating and verifying a .NET project) and involve standard, well-known development tooling.\n- [EXTERNAL_DOWNLOADS]: The skill fetches package updates and dependencies using dotnet restore from standard .NET package registries. It also references technical documentation and issue tracking from official Microsoft domains and GitHub repositories. These operations are conducted through trusted channels for the purpose of project migration.

migrate-dotnet10-to-dotnet11