The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data fetched from an external API, which could potentially contain malicious instructions or misleading content if the source were compromised.
Ingestion points: Genomic variant data and gene constraint metrics are ingested from the gnomAD GraphQL API via scripts/query_gnomad.py and Python examples in SKILL.md.
Boundary markers: The instructions do not include specific delimiters or warnings to the agent to ignore instructions that might be embedded in the scientific data retrieved from the API.
Capability inventory: The skill possesses the ability to execute Python scripts and perform network operations.
Sanitization: The skill lacks validation or sanitization routines for the JSON data returned by the Broad Institute's API.
[EXTERNAL_DOWNLOADS]: The skill interacts with the gnomAD GraphQL API at https://gnomad.broadinstitute.org/api. This is a well-known scientific service provided by the Broad Institute and is treated as a safe external reference for genomic research.
[COMMAND_EXECUTION]: The skill includes a Python script scripts/query_gnomad.py that is intended to be run by the agent to perform database lookups, which is a standard and intended behavior for this skill's research use case.

alterlab-gnomad