{ "schemaVersion": "1.0", "item": { "slug": "gemini-sub-agent", "name": "Gemini Sub-Agent", "source": "tencent", "type": "skill", "category": "开发工具", "sourceUrl": "https://clawhub.ai/tedo0626/gemini-sub-agent", "canonicalUrl": "https://clawhub.ai/tedo0626/gemini-sub-agent", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/gemini-sub-agent", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=gemini-sub-agent", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "SKILL.md", "scripts/setup.sh" ], "primaryDoc": "SKILL.md", "quickSetup": [ "Download the package from Yavira.", "Extract the archive and review SKILL.md first.", "Import or place the package into your OpenClaw setup." ], "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T16:55:25.780Z", "expiresAt": "2026-05-07T16:55:25.780Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentDisposition": "attachment; filename=\"network-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/gemini-sub-agent" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] }, "downloadPageUrl": "https://openagent3.xyz/downloads/gemini-sub-agent", "agentPageUrl": "https://openagent3.xyz/skills/gemini-sub-agent/agent", "manifestUrl": "https://openagent3.xyz/skills/gemini-sub-agent/agent.json", "briefUrl": "https://openagent3.xyz/skills/gemini-sub-agent/agent.md" }, "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "documentation": { "source": "clawhub", "primaryDoc": "SKILL.md", "sections": [ { "title": "gemini-sub-agent", "body": "Use Google Gemini (via subscription, $0 API cost) as a sub-agent inside OpenClaw workflows." }, { "title": "Setup", "body": "Run once on a fresh machine:\n\nbash skills/gemini-sub-agent/scripts/setup.sh\n\nThen authenticate (headless VPS flow):\n\nOn VPS: gcloud auth application-default login --no-browser → copy the --remote-bootstrap URL\nOn laptop: gcloud auth application-default login --remote-bootstrap=\"\" → copy the localhost:8085 output\nPaste localhost:8085 output back into VPS prompt\nComplete Gemini CLI OAuth: GOOGLE_GENAI_USE_GCA=true gemini -p \"hello\" → paste auth code\n\nCredentials are cached indefinitely and auto-refresh." }, { "title": "Available Models", "body": "ModelUse forgemini-3.1-pro-previewDefault — latest flagship, best reasoninggemini-2.5-proStable coding tasks, multi-file editsgemini-3-flash-previewFast summaries, quick Q&Agemini-3-pro-previewPrevious gen Pro, reliable fallbackgemini-2.5-flashLightweight fallback" }, { "title": "1. Simple text task (exec)", "body": "ask-gemini \"your prompt here\"\nask-gemini -m gemini-2.5-pro \"your prompt\"" }, { "title": "2. Pipe large content (long-context)", "body": "cat large_file.txt | ask-gemini \"summarize this in 5 bullets\"\ncat report.md | ask-gemini \"extract all action items\"" }, { "title": "3. Agentic coding (replaces Swift/Sonnet for 30-200 line tasks)", "body": "cd /path/to/project\nGOOGLE_GENAI_USE_GCA=true gemini -m gemini-2.5-pro -y -p \"write a script that...\"\n\n-y = yolo mode (auto-approves all file writes and shell commands). Gemini reads/writes files autonomously." }, { "title": "4. Pipe files directly into agentic session", "body": "cat existing_script.py | GOOGLE_GENAI_USE_GCA=true gemini -m gemini-2.5-pro -y -p \"refactor this to add error handling\"" }, { "title": "Routing Rules (when to use Gemini vs others)", "body": "TaskUseText, analysis, summarizationask-geminiMedium coding (30–200 lines)Gemini agentic (-y) instead of Swift/SonnetLarge file / long-contextask-gemini (1M token window)Web search / researchResearch agent (Grok) — Gemini has no web accessMulti-agent orchestrationOpenClaw native (sessions_spawn)Complex architecture / debuggingCodex (Opus) — escalate if Gemini fails twice" }, { "title": "Escalation", "body": "If Gemini returns wrong output twice on the same task → escalate to Swift (Sonnet) or Codex (Opus).\nLog the failure in failures/hot_antipatterns.md with the task type." }, { "title": "Scripts", "body": "scripts/setup.sh — Full install: gemini-cli + gcloud + ask-gemini wrapper\nscripts/ask-gemini — The wrapper script itself (copy to /usr/local/bin/)" } ], "body": "gemini-sub-agent\n\nUse Google Gemini (via subscription, $0 API cost) as a sub-agent inside OpenClaw workflows.\n\nSetup\n\nRun once on a fresh machine:\n\nbash skills/gemini-sub-agent/scripts/setup.sh\n\n\nThen authenticate (headless VPS flow):\n\nOn VPS: gcloud auth application-default login --no-browser → copy the --remote-bootstrap URL\nOn laptop: gcloud auth application-default login --remote-bootstrap=\"\" → copy the localhost:8085 output\nPaste localhost:8085 output back into VPS prompt\nComplete Gemini CLI OAuth: GOOGLE_GENAI_USE_GCA=true gemini -p \"hello\" → paste auth code\n\nCredentials are cached indefinitely and auto-refresh.\n\nAvailable Models\nModel\tUse for\ngemini-3.1-pro-preview\tDefault — latest flagship, best reasoning\ngemini-2.5-pro\tStable coding tasks, multi-file edits\ngemini-3-flash-preview\tFast summaries, quick Q&A\ngemini-3-pro-preview\tPrevious gen Pro, reliable fallback\ngemini-2.5-flash\tLightweight fallback\nUsage Patterns\n1. Simple text task (exec)\nask-gemini \"your prompt here\"\nask-gemini -m gemini-2.5-pro \"your prompt\"\n\n2. Pipe large content (long-context)\ncat large_file.txt | ask-gemini \"summarize this in 5 bullets\"\ncat report.md | ask-gemini \"extract all action items\"\n\n3. Agentic coding (replaces Swift/Sonnet for 30-200 line tasks)\ncd /path/to/project\nGOOGLE_GENAI_USE_GCA=true gemini -m gemini-2.5-pro -y -p \"write a script that...\"\n\n\n-y = yolo mode (auto-approves all file writes and shell commands). Gemini reads/writes files autonomously.\n\n4. Pipe files directly into agentic session\ncat existing_script.py | GOOGLE_GENAI_USE_GCA=true gemini -m gemini-2.5-pro -y -p \"refactor this to add error handling\"\n\nRouting Rules (when to use Gemini vs others)\nTask\tUse\nText, analysis, summarization\task-gemini\nMedium coding (30–200 lines)\tGemini agentic (-y) instead of Swift/Sonnet\nLarge file / long-context\task-gemini (1M token window)\nWeb search / research\tResearch agent (Grok) — Gemini has no web access\nMulti-agent orchestration\tOpenClaw native (sessions_spawn)\nComplex architecture / debugging\tCodex (Opus) — escalate if Gemini fails twice\nEscalation\n\nIf Gemini returns wrong output twice on the same task → escalate to Swift (Sonnet) or Codex (Opus). Log the failure in failures/hot_antipatterns.md with the task type.\n\nScripts\nscripts/setup.sh — Full install: gemini-cli + gcloud + ask-gemini wrapper\nscripts/ask-gemini — The wrapper script itself (copy to /usr/local/bin/)" }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/tedo0626/gemini-sub-agent", "publisherUrl": "https://clawhub.ai/tedo0626/gemini-sub-agent", "owner": "tedo0626", "version": "1.0.0", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/gemini-sub-agent", "downloadUrl": "https://openagent3.xyz/downloads/gemini-sub-agent", "agentUrl": "https://openagent3.xyz/skills/gemini-sub-agent/agent", "manifestUrl": "https://openagent3.xyz/skills/gemini-sub-agent/agent.json", "briefUrl": "https://openagent3.xyz/skills/gemini-sub-agent/agent.md" } }