The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from local .NET project files to determine configuration, which could potentially be manipulated to influence agent behavior.
Ingestion points: The skill reads .csproj, Program.cs, launchSettings.json, and appsettings.json files in Phase 0 to extract framework versions, ports, and health check paths.
Boundary markers: None. The instructions do not define delimiters or specific warnings to ignore instructions that might be embedded in these project files.
Capability inventory: The skill uses Shell access to execute aws CLI commands and docker operations, and Write access to generate infrastructure code and Dockerfiles.
Sanitization: Absent. Extracted values are used directly in shell commands (docker run) and generated code without validation.
[COMMAND_EXECUTION]: The skill uses shell tools to discover environment details and validate build artifacts.
Evidence: Executes aws ec2 describe-vpcs and aws ec2 describe-subnets to list network infrastructure for user selection.
Evidence: Executes docker build, docker run, and docker rm locally to verify that the generated Docker image starts correctly before recommending deployment.
[EXTERNAL_DOWNLOADS]: The skill references an external documentation source to ensure up-to-date AWS practices.
Evidence: Recommends the use of a Knowledge MCP server at https://knowledge-mcp.global.api.aws for looking up AWS resource schemas and documentation.

dotnet-aws-ecs