The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches documentation indices and content from mlflow.org. This aligns with the skill's stated purpose and targets the vendor's official documentation infrastructure.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from the MLflow documentation site, which represents an attack surface for indirect prompt injection if the source content were compromised.
Ingestion points: WebFetch operations in SKILL.md target mlflow.org/docs/latest/llms.txt and subsequent .md files.
Boundary markers: Absent; the prompts do not explicitly instruct the agent to ignore instructions found within the fetched documentation.
Capability inventory: Limited to network read operations via WebFetch; no file system write or shell execution capabilities are present.
Sanitization: Absent.

searching-mlflow-docs