busco-phylogeny

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests public third‑party content — e.g., the "Substitution Model Recommendation" step uses WebFetch to retrieve IQ‑TREE documentation and STEP 0A (scripts/query_ncbi_assemblies.py and download_ncbi_genomes.py) queries/downloads assemblies from NCBI — and the agent is expected to read and act on that untrusted web content to choose models and assemblies that materially influence subsequent tool execution.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly calls WebFetch at runtime to retrieve and inject IQ-TREE model documentation from https://iqtree.github.io/doc/Substitution-Models into the agent's context to drive substitution-model recommendations, meaning remote content would directly influence prompts/instructions.