The Agent Skills Directory

[SAFE]: The skill operates as a high-level instructional framework for critical thinking and data analysis. It does not contain any executable scripts, binary files, or network operations other than the explicitly mentioned use of a search tool.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied investment theses to generate search queries and analysis.
Ingestion points: The workflow begins by taking a user's narrative investment thesis in Step 1.
Boundary markers: No specific delimiters or safety instructions are defined to separate user input from the skill's logic.
Capability inventory: The skill utilizes the web_search tool to find counter-evidence.
Sanitization: The input is formatted into structured claims and adversarial search queries. While there is no explicit sanitization, the risk is minimal as the input is used for information retrieval rather than command execution.

@3182/thesis-falsifier