The Agent Skills Directory

Managed Agent Capabilities: The skill facilitates the use of the antigravity-preview-09-2026 agent, which possesses the capability to execute code (Bash, Python, Node.js), manage files, and access the web within a sandboxed environment. These features are part of the intended agentic functionality but require awareness of the capabilities available to the model.
Indirect Prompt Injection Surface: The skill allows for the ingestion of untrusted data through user inputs and external repository synchronization.
Ingestion points: Data enters the context via the input parameter in interaction creation and the sources field in agent environment configuration (SKILL.md).
Boundary markers: The instructions do not define specific delimiters or warnings to ignore instructions embedded within the processed data.
Capability inventory: The managed agents have access to shell command execution, file system operations, and network tools (SKILL.md).
Sanitization: There are no explicit instructions for sanitizing or validating external content before it is processed by the agent.
Package Installation: The skill references the installation of official libraries using pip install -U google-genai and npm install @google/genai. These are standard requirements for using the vendor's API.
Development Utility Commands: The migration guide provides a shell command using rg (ripgrep) to help analyze project scope. This is a standard developer tool used for static analysis of local files.

gemini-api-dev