The Agent Skills Directory

[SAFE]: The skill correctly uses the vendor-provided tabnine-coaching tool to manage coding standards, adhering to the principle of least privilege and purpose-built tool use.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of coding guidelines fetched via the tabnine-coaching MCP server. This establishes a surface for indirect prompt injection if the guidelines source is compromised.
Ingestion points: Guidelines retrieved from the get_guidelines tool in SKILL.md.
Boundary markers: Not present; the agent is instructed to fetch and apply guidelines directly.
Capability inventory: The skill is primarily focused on information retrieval and code review feedback.
Sanitization: No explicit validation or filtering of the retrieved content is defined.

coding-guidelines