The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses dynamic context injection (the ! command syntax) to read project configuration files like .production-grade.yaml and .forgewright/codebase-context.md to initialize the agent's environment. It also provides shell and PowerShell scripts for executing platform-specific build tools such as xcodebuild, gradlew, and Unity's command-line interface.
[REMOTE_CODE_EXECUTION]: The skill documents a HotfixManager implementation in C# that allows a game to download and apply updates at runtime. This functionality is a standard part of game maintenance and is provided as a reference implementation for users.
[PROMPT_INJECTION]: The skill architecture is susceptible to indirect prompt injection through the ingestion of repository files and workflow configurations.
Ingestion points: Files located at .forgewright/project-profile.json and .github/workflows/ are read to determine project state.
Boundary markers: No explicit delimiter-based sanitization is implemented in the provided code snippets.
Capability inventory: The skill includes extensive capabilities for file system modification, command execution, and network communication via the generated build and release scripts.
Sanitization: Security is maintained through standard cryptographic hash verification (SHA256) within the provided hotfix delivery template.

build-release-engineer