The Agent Skills Directory

[SAFE]: No malicious behaviors such as prompt injection, data exfiltration, or obfuscation were detected. The skill's permissions are appropriate for its research-oriented functionality.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to ingest and process external academic sources. Ingestion points: Research material and manuscript claims entering through network:http and file:read (SKILL.md). Boundary markers: The skill does not explicitly define delimiters or instructions to ignore commands within processed source material. Capability inventory: The skill utilizes network:http, file:read, and file:write permissions. Sanitization: No sanitization or validation of external source content is specified.

nonfiction-research