Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →lingtai-ai/preset-skill-gemini
Official-source-led manual for the TUI `gemini` template.
Passed 21 of 23 test cases with the skill, and 5 of 23 without it.
and 18 more in the full report
17 fixed4 passed either way1 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 17 test cases the skill fixed. Same model, same prompt, with and without the skill.
When revising the Gemini preset constructor in tui/internal/preset/preset.go, which exact three fields should be modified?
| Checked | Without skill | With skill |
|---|---|---|
| specifies model as an editable constructor field | Passed | Passed |
| specifies credential as an editable constructor field | Passed | Passed |
| specifies vision capability as an editable constructor field | Missed | Passed |
The instructions your model reads when the skill loads.
Use this child for the named built-in gemini preset. geminiPreset in tui/internal/preset/preset.go:1332 uses Google's native gemini adapter, the stable gemini-3.8-flash default, GEMINI_API_KEY, web_search, skills, and a native vision capability. gemini-3-flash-preview was the previous preview default. It has no base_url or OpenAI-compatibility override.
Native inline image input is distinct from the explicit LingTai vision
capability; both use the shipped Gemini provider rather than a fallback.
The explicit LingTai vision capability is a separate tool path from native
multimodality.
Read Google's official Gemini model guide and image understanding guide. For the actual account-served list, call the official Gemini API models.list endpoint, GET https://generativelanguage.googleapis.com/v1beta/models, or use the equivalent official SDK method; follow pagination and filter for models supporting the operation used by this adapter. This is a native provider catalog, not a gateway alias. Keep preview/stability and image-input facts separate, and do not infer a LingTai MCP from native multimodality.
Start at geminiPreset in tui/internal/preset/preset.go. Gemini intentionally has no providerModels picker or modelHasVision entry in tui/internal/tui/preset_editor.go, so revise only the constructor's model, credential, and vision capability unless that design changes. If capability display changes, inspect the fixed mandatoryCapRow rendering; there is no Gemini base_url surface. Follow tui/CONTRACT.md for free-text providers.
Prepare an evidence-bound manifest and explicit input, then run lingtai-tui presets revise --manifest PATH --input PATH --mode dry-run|check|apply [--output-dir PATH]. Review the JSON plan, use dry-run/check before apply, and apply only to a new explicit output directory. revision.go validates hashes, route bindings, and evidence and preserves unowned bytes. This amendment
test cases passed, vs 5 / 23 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.