v0.2.16

Added

• List gemini-3.5-flash-lite for text requests through Chat Completions and native Gemini content generation.
• Gemini 3.8 Flash (gemini-3.8-flash) is now listed as an available text-only Gemini model with function calling.

v0.2.15

Added

• Document Gemini 3.7 Flash native streaming and OpenAI-compatible streaming with terminal usage verification.

v0.2.14

Changed

• Document Gemini authentication, request, result verification, and stop boundaries in the primary recipe, with compatibility protocols loaded only when required.

v0.2.13

Changed

• Complete marketplace distribution for the RunAPI Gemini skill.

v0.2.12

Added

• Document gemini-3.6-flash for OpenAI-compatible Chat and Gemini native content streaming.

v0.2.11

Removed

• Remove gemini-3-pro-preview from the supported Gemini model list.
  Migration: Choose another currently supported model ID; no replacement alias is provided.

v0.2.10

Changed

• Publish the Gemini skill v0.2.10 release with refreshed RunAPI routing metadata.

v0.2.9

Changed

• Supported models table simplified to model + capabilities; protocol support (OpenAI-compatible + Gemini contents) is described in the request-styles section.

v0.2.8

Changed

• Supported models table is now generated from the RunAPI catalog.
• Corrected gemini-3.5-flash Chat Completions support and refreshed model ordering.

v0.2.7

Changed

• Clarify Gemini contents support and native versus bridged streaming behavior.