The Agent Skills Directory

[SAFE]: The skill is composed entirely of markdown documentation and reference files. There is no executable code, such as Python scripts, shell commands, or configuration for automated tools included in the skill package.\n- [SAFE]: External links point to a well-known service (Amazon) to provide references for the source literature. These links include the author's affiliate tag (wondelai00-20), which is consistent with the skill's author metadata and represents standard vendor resource usage.\n- [SAFE]: The instructions focus on providing a reasoning framework for the AI agent and do not contain patterns associated with prompt injection, credential harvesting, or data exfiltration.\n- [SAFE]: Risk of indirect prompt injection is assessed. Ingestion point: The skill is designed to process user-provided strategy documents and pitch decks (documented in SKILL.md). Boundary markers: None explicitly defined. Capability inventory: No executable scripts, subprocess calls, tools, network operations, or file-write capabilities are defined in any file. Sanitization: No explicit data sanitization is performed. Due to the complete absence of dangerous capabilities, this attack surface is considered safe.

good-strategy-bad-strategy