The Agent Skills Directory

[SAFE]: The skill consists of markdown-based instructional content, hypothesis frameworks, and documentation templates. It does not contain executable scripts, shell commands, or obfuscated payloads.
[EXTERNAL_DOWNLOADS]: The skill provides links to several well-known third-party A/B testing calculators and tools, including those from Evan Miller, Optimizely, and VWO. These are established industry resources and are documented here as safe references for the intended functionality.
[INDIRECT_PROMPT_INJECTION]: The skill contains instructions to ingest data from local project files to gather product marketing context.
Ingestion points: Reads files such as .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md if they exist.
Boundary markers: No explicit delimiters or warnings to ignore embedded instructions are present when reading these context files.
Capability inventory: The skill is limited to text generation and analysis within the agent's conversation. It does not have access to subprocess execution, file writing, or network transmission tools.
Sanitization: No specific validation or escaping is applied to the content of the ingested files before they are processed by the agent.

ab-testing