The Agent Skills Directory

[SAFE]: The skill implementation in SKILL.md and scripts/rna_velocity_workflow.py follows established bioinformatics protocols for RNA velocity estimation. It relies on legitimate, open-source libraries such as scvelo, scanpy, and anndata for all computational tasks.
[SAFE]: All file operations are restricted to local data ingestion (e.g., reading .loom and .h5ad files) and saving results to a local output directory. There are no network operations that could lead to data exfiltration or unauthorized remote communication.
[SAFE]: No obfuscation techniques, hidden commands, or suspicious encoding patterns (such as Base64 or hex-encoded payloads) were detected across the skill files. The technical documentation in references/velocity_models.md accurately describes the mathematical and computational methods used.
[SAFE]: The skill does not exhibit any persistence mechanisms, privilege escalation attempts, or prompt injection vectors. It acts as a specialized data analysis tool for life sciences researchers without exceeding its intended scope.

scvelo