The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill architecture involves the agent processing user-supplied manifest files (xl1.dapp.json and xl1.deploy.json) and external event data (Event Kit wakes). This ingestion of external, potentially untrusted data creates a surface where malicious instructions could be embedded to influence the agent's behavior.
Ingestion points: Repository manifest files and event streams from Event Kit.
Boundary markers: The framework utilizes schema validation via the xl1-dapp validate command, though it does not explicitly document boundary markers for LLM context protection.
Capability inventory: The associated CLI tool can execute build and test scripts and manage deployment locks.
Sanitization: Employs neutral planning phases and schema-only target contracts to limit unvalidated I/O.
[DYNAMIC_EXECUTION]: The xl1-dapp CLI tool is designed to execute scripts defined within the project manifest during the build and test phases. This mechanism allows the execution of code defined in configuration files, which is a standard but noteworthy execution surface for development tools.
[EXTERNAL_DOWNLOADS]: The documentation identifies a comprehensive suite of NPM packages under the @xyo-network scope required for building dApp Kit applications. These are recognized as legitimate vendor resources for the framework described.
[COMMAND_EXECUTION]: The skill instructs the use of the xl1-dapp and xl1-dapp-vercel CLI tools to perform project management tasks, including initialization, validation, building, and deployment orchestration.

xl1-dapp-kit