The Agent Skills Directory

[SAFE]: The skill's instructions focus on organizational and research-oriented logic without including any malicious code, obfuscation, or unauthorized data access commands.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it instructs the agent to ingest external data from research papers, benchmarks, and technical reports (Phase 1.2). This risk factor is assessed as safe because the skill lacks executable capabilities or system-level access that could be leveraged for an attack. Boundary markers for external content are not explicitly defined.

computer-scientist