The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from user project files, creating an attack surface for indirect prompt injection.
Ingestion points: The agent reads and interprets content from .csproj, .sln, .slnx, Dockerfile, and C# source files (SKILL.md, Step 1).
Boundary markers: The workflow explicitly instructs the agent to treat the provided reference documents as authoritative and to avoid unauthorized external web searches or unnecessary filesystem exploration (SKILL.md, Workflow).
Capability inventory: The skill utilizes dotnet CLI tools for building, testing, and restoring packages, and it has the capability to modify project and source files (SKILL.md, Steps 1-6).
Sanitization: No specific sanitization or escaping mechanisms for the processed source code are described in the instructions.
[COMMAND_EXECUTION]: The skill requires the execution of shell commands to establish baselines, perform restores, and verify the migration.
Execution patterns: The agent runs dotnet --version, dotnet build, dotnet restore, and dotnet test as part of its core workflow (SKILL.md).
Infrastructure updates: The skill guides the user through updating Dockerfiles, which may involve executing apt-get commands within container environments (references/containers-interop-dotnet9to10.md).

migrate-dotnet9-to-dotnet10