The Agent Skills Directory

[SAFE]: The skill's behavior is consistent with its stated purpose of applying the Tetlock superforecasting methodology. No evidence of malicious intent, data exfiltration, or unauthorized access was found.
[COMMAND_EXECUTION]: The skill executes a simple bash command to check the configuration of the execution environment. This is a benign operation used to determine whether to use parallel agent teams or sequential agent calls.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests untrusted data from user-provided theses and external web sources via WebSearch and WebFetch. * Ingestion points: User-provided arguments and web content retrieved by sub-agents for market research and base rate data. * Boundary markers: The skill uses markdown placeholders and structured output templates but does not implement explicit delimiters or instructions to ignore embedded commands in the fetched web data. * Capability inventory: The skill can spawn sub-agents, perform web operations, and write analysis results to files. * Sanitization: There is no explicit sanitization or validation logic for the external content before it is processed by the sub-agents or the synthesis lead.

tetlock