The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected in the skill instructions or metadata.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided paper metadata (titles, abstracts) which constitutes an indirect prompt injection surface. However, because the skill has no dangerous capabilities (such as shell access, network requests, or file writing), the risk is negligible.
[INGESTION_POINTS]: Untrusted data enters the agent context through the seed_papers variable in prompt.md.
[BOUNDARY_MARKERS]: No explicit boundary markers or 'ignore embedded instructions' warnings are present to isolate user-provided data from system instructions.
[CAPABILITY_INVENTORY]: The skill possesses no capabilities for command execution, network operations, or persistent storage.
[SANITIZATION]: There is no explicit sanitization or validation of input data described in the instructions.

seed-selection