The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill functions by taking user-supplied research topics and fields and interpolating them into internal prompts for drafting academic content. While there are no strict boundary markers to isolate untrusted data, the skill possesses no access to tools, network requests, or system execution capabilities, rendering the vulnerability surface non-exploitable.
Ingestion points: User-provided variables for Topic, Field, and Scope in the SKILL.md menu and templates/prompt-main.md patterns.
Boundary markers: Absent for user-supplied interpolation values.
Capability inventory: No file system writes, network operations, or shell command executions are available to the skill.
Sanitization: The skill does not validate or sanitize user input before inclusion in the final output generation instructions.
[EXTERNAL_DOWNLOADS]: The skill provides informational links to established academic services such as Google Scholar (scholar.google.com), PubMed, and ERIC (eric.ed.gov). These are intended for manual user verification of citations and are not used for automated code execution or data transfer.

researcher-pro