The Agent Skills Directory

[SAFE]: The skill consists of markdown-based instructions and metadata defining an investment framework. No executable scripts, system commands, or code dependencies are present across the provided files.
[EXTERNAL_DOWNLOADS]: The skill references external content on Spotify, YouTube, and the official a16zcrypto.com domain. These target well-known platforms and the official organization website, which are considered safe sources.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-provided market data and project contexts.
Ingestion points: Processes asset context, on-chain data, and research data (SKILL.md).
Boundary markers: Instructions request the separation of facts, assumptions, and judgment, but do not provide explicit structural delimiters for untrusted data.
Capability inventory: The skill has no code execution, file system writing, or network transmission capabilities.
Sanitization: No specific input sanitization is described. However, the lack of dangerous capabilities makes this surface low-risk.
[NO_CODE]: The skill is composed entirely of informational markdown and YAML configuration, presenting no direct execution risk.

a16z-crypto