The Agent Skills Directory

[SAFE]: No security issues detected. The skill is composed entirely of markdown instructions and YAML configuration for research paper analysis.
[PROMPT_INJECTION]: The skill was evaluated for indirect prompt injection surfaces due to its ingestion of user-provided data.
Ingestion points: User-provided seed_papers and context strings defined in SKILL.md.
Boundary markers: Absent; inputs are interpolated directly into the strategist prompt in prompt.md without specific delimiters.
Capability inventory: None; the skill is limited to text-based reasoning and lacks file system access, network operations, or subprocess execution capabilities.
Sanitization: None; the skill depends on the underlying model's internal safety guardrails for input processing.

seed-selection