Stats
Actions
Tags
How this skill is triggered — by the user, by Claude, or both
Slash command
/gemini:gemini-result-handlingThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Standards for presenting Gemini CLI output back to the user.
Standards for presenting Gemini CLI output back to the user.
Preserve verbatim. Return Gemini's verdict, summary, findings, file paths, and line numbers exactly as received.
Severity ordering. Present findings ordered by severity: critical → error → warning → info.
File references. Maintain exact file paths and line numbers (e.g., src/auth.ts:42). Do not resolve, abbreviate, or strip these references.
Stop after findings. After presenting review findings, STOP. Do not:
The user must explicitly request follow-up action.
No substitution. Never substitute Claude-side analysis for incomplete or failed Gemini runs. If Gemini's run was incomplete:
Empty results. When Gemini reports no findings, state "No issues found" explicitly. Do not invent issues or add caveats.
When presenting results from /gemini:result:
--resumenpx claudepluginhub jiachengxu/gemini-plugin-cc --plugin geminiInternal guidance for presenting Gemini helper output back to the user
Parses Gemini CLI JSON and stream-JSON outputs for responses, token stats, tool calls, errors. Use in automation pipelines processing headless CLI results.
Delegates tasks to Gemini CLI for large-context analysis like broad codebase reviews or long-document processing. Activates on explicit requests such as 'use gemini' or 'delegate to gemini'.