The Agent Skills Directory

[SAFE]: The skill operates as a technical interface for biological sequence analysis. It does not contain any malicious instructions, obfuscation, or persistence mechanisms. All network activity is limited to the official vendor API (genomicintelligence.ai) and the trusted Ensembl genomics platform (ensembl.org).
[INDIRECT_PROMPT_INJECTION]: The skill provides an interface for processing DNA sequence data and genomic identifiers.
Ingestion points: Biological sequences, gene symbols, and genomic coordinates are ingested via tool parameters such as sequence, gene, and region (SKILL.md, references/mcp.md).
Boundary markers: Data is expected to adhere to biological sequence formats (nucleotide alphabet), which naturally delimits the input from command-like instructions (references/tasks.md).
Capability inventory: The skill's primary capability is performing networked model inference via requests.post and requests.get to the vendor's API (references/api-and-auth.md).
Sanitization: The skill documents that the backend API performs strict alphabet validation (A, C, G, T, N) and length checks, which serves as a robust filter for non-genomic content (references/tasks.md).

genomic-intelligence