The Agent Skills Directory

[SAFE]: The skill serves as a development guide and instructional framework for Solana applications. It contains no executable malicious code, obfuscation, or unauthorized data access attempts.
[EXTERNAL_DOWNLOADS]: The skill recommends using standard Solana ecosystem libraries such as @solana/kit, @solana/client, Anchor, Pinocchio, and LiteSVM. These are reputable, well-known tools within the blockchain development community.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-initiated Solana development tasks.
Ingestion points: User-provided requests for UI work, program development, or testing (SKILL.md).
Boundary markers: Not explicitly defined in the provided instructions.
Capability inventory: Code generation, diff application, and execution of build/test commands (SKILL.md).
Sanitization: Not applicable as the skill primarily generates structured code based on development best practices. This represents a standard functional surface for a developer-oriented agent.

solana-dev