About gemini-runner
A skill that dispatches a caller-scoped prompt through the local Antigravity CLI, `agy`, while following an approved Gemini model snapshot. It checks the serving model and route outcome rather than treating process completion as proof of success.
Key features
- 01Run prompts through local `agy` CLI
- 02Use an approved Gemini model snapshot
- 03Check provider and serving-model receipt
- 04Report fallback and malformed output
- 05Resume shared conversations when configured
Use cases
- 01Dispatching a scoped task to Gemini
- 02Verifying which provider answered
- 03Handling unavailable or unverified model routes
- 04Continuing a shared CLI conversation
Install
Copy the command and run it in a terminal.
npx skills add https:/ / github.com/ robsonrung/ rar-skills --skill gemini-runnerAfter installing, type/gemini-runnerin Claude Code to confirm it loaded.
Bundled files
References and scripts installed along with SKILL.md.
agents/openai.yamlreferences/command-reference.mdcontinuation.mdruntime-reference.mdscripts/run_gemini.pytests/test_run_gemini.pyDetails
From the project READMEGemini Runner
Use shared/references/model-preview.md to choose the branch first. Deterministic checks need no model selection. Actual dispatch uses one concrete approved snapshot; nested calls consume it without re-resolution or extra workers.
Execute the caller's scoped prompt through the local CLI. Shared roles, envelope keys, and result handling live in shared/references/runner-common.md. Preserve seat fidelity: name the provider that actually answered, and report unavailable or unverified seats honestly.
Native routing
Read shared/references/host-model-execution.md before choosing an external route. If the current host can dispatch the exact selected Gemini model with the required effort, isolation, and receipt policy in a native subagent or task thread, use that route. agents/openai.yaml is host UI metadata; it does not dispatch a native model.
Invocation
Set SKILL_DIR to this loaded skill directory in the same shell call:
SKILL_DIR="<absolute path of this skill directory>";
python3 "$SKILL_DIR/scripts/run_gemini.py" "<scoped prompt>" \
--model <approved-model-id> --disable-fallback --json
Copy the selected model label, role, tool mode, receipt policy, and supported limits from the snapshot. This runner has no effort flag or native effort control. A route requiring unsupported control stays blocked. Nested calls use the saved options without resolving defaults again. The example does not grant source sharing or tool authority.
Runtime boundaries
- Use the local authenticated Antigravity CLI,
agy, in print mode. - The user configures the model in agy.
--modeland output-format flags are request labels or prompt hints, not native model or format controls. - Permission checks follow local configuration. Prompt and file context may be sent to the configured provider.
- Analysis roles use a read-only prompt overlay, not an enforced sandbox. Isolate the workspace for untrusted input;
--restrict-toolsalone is not a security boundary. - Missing-CLI fallback is labelled and described in the runtime reference. Approved routes use
--disable-fallback; an unverified model requires the caller's explicit receipt policy. - Native output has no session ID.
--agy-continueresumes the latest shared conversation and must not run concurrently against the same home.
Load by need
- references/command-reference.md: flags, examples, native mappings, and return codes when composing a non-basic invocation.
- references/runtime-reference.md: authentication, fallback conditions, configuration, and extended envelope fields when configuring a route or interpreting its result.
- references/continuation.md: only when resuming a session or managing a background job.
Result
Read agent_message and the wrapper envelope. Check success, serving-model receipt, effective runner/provider, and any fallback or malformed-output status. A completed process alone does not prove the requested task or model was used. Report touched files when the caller authorized edits.