{"id":"2c1f3ff0-6f18-463e-aa42-e0174f5c05bc","shortId":"64EdTz","kind":"skill","title":"create-gemini-agent","tagline":"Create a custom Gemini agent by defining its system prompt, tools, and configuration in a markdown file.","description":"# Subagents\n\nSubagents are specialized agents that operate within your main Gemini CLI\nsession. They are designed to handle specific, complex tasks—like deep codebase\nanalysis, documentation lookup, or domain-specific reasoning—without cluttering\nthe main agent's context or toolset.\n\n## What are subagents?\n\nSubagents are \"specialists\" that the main Gemini agent can hire for a specific\njob.\n\n- **Focused context:** Each subagent has its own system prompt and persona.\n- **Specialized tools:** Subagents can have a restricted or specialized set of\n  tools.\n- **Independent context window:** Interactions with a subagent happen in a\n  separate context loop, which saves tokens in your main conversation history.\n\nSubagents are exposed to the main agent as a tool of the same name. When the\nmain agent calls the tool, it delegates the task to the subagent. Once the\nsubagent completes its task, it reports back to the main agent with its\nfindings.\n\n## How to use subagents\n\nYou can use subagents through automatic delegation or by explicitly forcing them\nin your prompt.\n\n### Automatic delegation\n\nGemini CLI's main agent is instructed to use specialized subagents when a task\nmatches their expertise. For example, if you ask \"How does the auth system\nwork?\", the main agent may decide to call the `codebase_investigator` subagent\nto perform the research.\n\n### Forcing a subagent (@ syntax)\n\nYou can explicitly direct a task to a specific subagent by using the `@` symbol\nfollowed by the subagent's name at the beginning of your prompt. This is useful\nwhen you want to bypass the main agent's decision-making and go straight to a\nspecialist.\n\n**Example:**\n\n```bash\n@codebase_investigator Map out the relationship between the AgentRegistry and the LocalAgentExecutor.\n```\n\nWhen you use the `@` syntax, the CLI injects a system note that nudges the\nprimary model to use that specific subagent tool immediately.\n\n## Creating custom subagents\n\nYou can create your own subagents to automate specific workflows or enforce\nspecific personas.\n\n### Agent definition files\n\nCustom agents are defined as Markdown files (`.md`) with YAML frontmatter. You\ncan place them in:\n\n1.  **Project-level:** `.gemini/agents/*.md` (Shared with your team)\n2.  **User-level:** `~/.gemini/agents/*.md` (Personal agents)\n\n### File format\n\nThe file **MUST** start with YAML frontmatter enclosed in triple-dashes `---`.\nThe body of the markdown file becomes the agent's **System Prompt**.\n\n**Example: `.gemini/agents/security-auditor.md`**\n\n```markdown\n---\nname: security-auditor\ndescription: Specialized in finding security vulnerabilities in code.\nkind: local\ntools:\n  - read_file\n  - grep_search\nmodel: gemini-3-flash-preview\ntemperature: 0.2\nmax_turns: 10\n---\n\nYou are a ruthless Security Auditor. Your job is to analyze code for potential\nvulnerabilities.\n\nFocus on:\n\n1.  SQL Injection\n2.  XSS (Cross-Site Scripting)\n3.  Hardcoded credentials\n4.  Unsafe file operations\n\nWhen you find a vulnerability, explain it clearly and suggest a fix. Do not fix\nit yourself; just report it.\n```\n\n### Configuration schema\n\n| Field          | Type   | Required | Description                                                                                                                                                                                                   |\n| :------------- | :----- | :------- | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |\n| `name`         | string | Yes      | Unique identifier (slug) used as the tool name for the agent. Only lowercase letters, numbers, hyphens, and underscores.                                                                                      |\n| `description`  | string | Yes      | Short description of what the agent does. This is visible to the main agent to help it decide when to call this subagent.                                                                                     |\n| `kind`         | string | No       | `local` (default) or `remote`.                                                                                                                                                                                |\n| `tools`        | array  | No       | List of tool names this agent can use. Supports wildcards: `*` (all tools), `mcp_*` (all MCP tools), `mcp_server_*` (all tools from a server). **If omitted, it inherits all tools from the parent session.** |\n| `mcpServers`   | object | No       | Configuration for inline Model Context Protocol (MCP) servers isolated to this specific agent.                                                                                                                |\n| `model`        | string | No       | Specific model to use (for example, `gemini-3-preview`). Defaults to `inherit` (uses the main session model).                                                                                                 |\n| `temperature`  | number | No       | Model temperature (0.0 - 2.0). Defaults to `1`.                                                                                                                                                               |\n| `max_turns`    | number | No       | Maximum number of conversation turns allowed for this agent before it must return. Defaults to `30`.                                                                                                          |\n| `timeout_mins` | number | No       | Maximum execution time in minutes. Defaults to `10`.                                                                                                                                                          |\n\n### Tool wildcards\n\nWhen defining `tools` for a subagent, you can use wildcards to quickly grant\naccess to groups of tools:\n\n- `*`: Grant access to all available built-in and discovered tools.\n- `mcp_*`: Grant access to all tools from all connected MCP servers.\n- `mcp_my-server_*`: Grant access to all tools from a specific MCP server named\n  `my-server`.\n\n### Isolation and recursion protection\n\nEach subagent runs in its own isolated context loop. This means:\n\n- **Independent history:** The subagent's conversation history does not bloat\n  the main agent's context.\n- **Isolated tools:** The subagent only has access to the tools you explicitly\n  grant it.\n- **Recursion protection:** To prevent infinite loops and excessive token usage,\n  subagents **cannot** call other subagents. If a subagent is granted the `*`\n  tool wildcard, it will still be unable to see or invoke other agents.\n\n## Subagent tool isolation\n\nSubagent tool isolation moves Gemini CLI away from a single global tool\nregistry. By providing isolated execution environments, you can ensure that\nsubagents only interact with the parts of the system they are designed for. This\nprevents unintended side effects, improves reliability by avoiding state\ncontamination, and enables fine-grained permission control.\n\nWith this feature, you can:\n\n- **Specify tool access:** Define exactly which tools an agent can access using\n  a `tools` list in the agent definition.\n- **Define inline MCP servers:** Configure Model Context Protocol (MCP) servers\n  (which provide a standardized way to connect AI models to external tools and\n  data sources) directly in the subagent's markdown frontmatter, isolating them\n  to that specific agent.\n- **Maintain state isolation:** Ensure that subagents only interact with their\n  own set of tools and servers, preventing side effects and state contamination.\n- **Apply subagent-specific policies:** Enforce granular rules in your\n  [Policy Engine](/docs/reference/policy-engine) TOML configuration based on the\n  executing subagent's name.\n\n### Configuring isolated tools and servers\n\nYou can configure tool isolation for a subagent by updating its markdown\nfrontmatter. This lets you explicitly state which tools the subagent can use,\nrather than relying on the global registry.\n\nAdd an `mcpServers` object to define inline MCP servers that are unique to the\nagent.\n\n**Example:**\n\n```yaml\n---\nname: my-isolated-agent\ntools:\n  - grep_search\n  - read_file\nmcpServers:\n  my-custom-server:\n    command: 'node'\n    args: ['path/to/server.js']\n---\n```\n\n### Subagent-specific policies\n\nYou can enforce fine-grained control over subagents using the\n[Policy Engine's](/docs/reference/policy-engine) TOML configuration. This allows\nyou to grant or restrict permissions specifically for an agent, without\naffecting the rest of your CLI session.\n\nTo restrict a policy rule to a specific subagent, add the `subagent` property to\nthe `[[rules]]` block in your `policy.toml` file.\n\n**Example:**\n\n```toml\n[[rules]]\nname = \"Allow pr-creator to push code\"\nsubagent = \"pr-creator\"\ndescription = \"Permit pr-creator to push branches automatically.\"\naction = \"allow\"\ntoolName = \"run_shell_command\"\ncommandPrefix = \"git push\"\n```\n\nIn this configuration, the policy rule only triggers if the executing subagent's\nname matches `pr-creator`. Rules without the `subagent` property apply\nuniversally to all agents.\n\n## Managing subagents\n\nYou can manage subagents interactively using the `/agents` command or\npersistently via `settings.json`.\n\n### Interactive management (/agents)\n\nIf you are in an interactive CLI session, you can use the `/agents` command to\nmanage subagents without editing configuration files manually. This is the\nrecommended way to quickly enable, disable, or re-configure agents on the fly.\n\nFor a full list of sub-commands and usage, see the\n[`/agents` command reference](/docs/reference/commands#agents).\n\n### Persistent configuration (settings.json)\n\nWhile the `/agents` command and agent definition files provide a starting point,\nyou can use `settings.json` for global, persistent overrides. This is useful for\nenforcing specific models or execution limits across all sessions.\n\n#### `agents.overrides`\n\nUse this to enable or disable specific agents or override their run\nconfigurations.\n\n```json\n{\n  \"agents\": {\n    \"overrides\": {\n      \"security-auditor\": {\n        \"enabled\": false,\n        \"runConfig\": {\n          \"maxTurns\": 20,\n          \"maxTimeMinutes\": 10\n        }\n      }\n    }\n  }\n}\n```\n\n#### `modelConfigs.overrides`\n\nYou can target specific subagents with custom model settings (like system\ninstruction prefixes or specific safety settings) using the `overrideScope`\nfield.\n\n```json\n{\n  \"modelConfigs\": {\n    \"overrides\": [\n      {\n        \"match\": { \"overrideScope\": \"security-auditor\" },\n        \"modelConfig\": {\n          \"generateContentConfig\": {\n            \"temperature\": 0.1\n          }\n        }\n      }\n    ]\n  }\n}\n```\n\n#### Safety policies (TOML)\n\nYou can restrict access to specific subagents using the CLI's **Policy Engine**.\nSubagents are treated as virtual tool names for policy matching purposes.\n\nTo govern access to a subagent, create a `.toml` file in your policy directory\n(e.g., `~/.gemini/policies/`):\n\n```toml\n[[rule]]\ntoolName = \"codebase_investigator\"\ndecision = \"deny\"\ndeny_message = \"Deep codebase analysis is restricted for this session.\"\n```\n\nFor more information on setting up fine-grained safety guardrails, see the\n[Policy Engine reference](/docs/reference/policy-engine#special-syntax-for-subagents).\n\n### Optimizing your subagent\n\nThe main agent's system prompt encourages it to use an expert subagent when one\nis available. It decides whether an agent is a relevant expert based on the\nagent's description. You can improve the reliability with which an agent is used\nby updating the description to more clearly indicate:\n\n- Its area of expertise.\n- When it should be used.\n- Some example scenarios.\n\nFor example, the following subagent description should be called fairly\nconsistently for Git operations.\n\n> Git expert agent which should be used for all local and remote git operations.\n> For example:\n>\n> - Making commits\n> - Searching for regressions with bisect\n> - Interacting with source control and issues providers such as GitHub.\n\nIf you need to further tune your subagent, you can do so by selecting the model\nto optimize for with `/model` and then asking the model why it does not think\nthat your subagent was called with a specific prompt and the given description.\n\n## Remote subagents (Agent2Agent)\n\nGemini CLI can also delegate tasks to remote subagents using the Agent-to-Agent\n(A2A) protocol.\n\nSee the [Remote Subagents documentation](/docs/core/remote-agents) for detailed\nconfiguration, authentication, and usage instructions.\n\n## Extension subagents\n\nExtensions can bundle and distribute subagents. See the\n[Extensions documentation](/docs/extensions#subagents) for details on how\nto package agents within an extension.\n\n## Disabling subagents\n\nSubagents are enabled by default. To disable them, set `enableAgents` to `false`\nin your `settings.json`:\n\n```json\n{\n  \"experimental\": { \"enableAgents\": false }\n}\n```","tags":["create","gemini","agent","jup","andrader","agent-skills","agentic-ai","agents","ai-agents","cli","installer","manager"],"capabilities":["skill","source-andrader","skill-create-gemini-agent","topic-agent","topic-agent-skills","topic-agentic-ai","topic-agents","topic-ai-agents","topic-cli","topic-installer","topic-manager","topic-skills"],"categories":["jup"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/andrader/jup/create-gemini-agent","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add andrader/jup","source_repo":"https://github.com/andrader/jup","install_from":"skills.sh"}},"qualityScore":"0.455","qualityRationale":"deterministic score 0.46 from registry signals: · indexed on github topic:agent-skills · 10 github stars · SKILL.md body (12,988 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-04-24T07:03:32.449Z","embedding":null,"createdAt":"2026-04-23T13:04:10.545Z","updatedAt":"2026-04-24T07:03:32.449Z","lastSeenAt":"2026-04-24T07:03:32.449Z","tsv":"'-3':424,608 '/.gemini/agents':370 '/.gemini/policies':1368 '/agents':1164,1172,1185,1224,1234 '/docs/core/remote-agents':1590 '/docs/extensions':1610 '/docs/reference/commands':1227 '/docs/reference/policy-engine':950,1050,1402 '/model':1541 '0.0':623 '0.1':1325 '0.2':429 '1':356,450,627 '10':432,659,1291 '2':366,453 '2.0':624 '20':1289 '3':459 '30':647 '4':462 'a2a':1583 'access':675,681,693,707,756,861,869,1332,1355 'across':1262 'action':1118 'add':996,1082 'affect':1066 'agent':4,9,26,58,73,130,141,164,193,219,272,337,341,373,396,505,521,529,554,597,640,747,797,867,876,915,1010,1017,1064,1154,1208,1228,1237,1273,1280,1413,1432,1440,1451,1490,1580,1582,1618 'agent-to-ag':1579 'agent2agent':1567 'agentregistri':293 'agents.overrides':1265 'ai':895 'allow':637,1054,1098,1119 'also':1571 'analysi':46,1380 'analyz':443 'appli':938,1150 'area':1463 'arg':1030 'array':547 'ask':210,1544 'auditor':406,438,1284,1321 'auth':214 'authent':1594 'autom':330 'automat':177,187,1117 'avail':684,1427 'avoid':844 'away':807 'back':160 'base':953,1437 'bash':284 'becom':394 'begin':258 'bisect':1510 'bloat':744 'block':1089 'bodi':389 'branch':1116 'built':686 'built-in':685 'bundl':1602 'bypass':269 'call':142,223,536,776,1482,1556 'cannot':775 'clear':473,1460 'cli':33,190,303,806,1071,1179,1338,1569 'clutter':55 'code':414,444,1104 'codebas':45,225,285,1372,1379 'command':1028,1123,1165,1186,1219,1225,1235 'commandprefix':1124 'commit':1505 'complet':155 'complex':41 'configur':17,486,585,882,952,960,967,1052,1129,1192,1207,1230,1278,1593 'connect':699,894 'consist':1484 'contamin':846,937 'context':60,81,104,114,589,731,749,884 'control':853,1042,1514 'convers':122,635,740 'creat':2,5,320,325,1359 'create-gemini-ag':1 'creator':1101,1108,1113,1144 'credenti':461 'cross':456 'cross-sit':455 'custom':7,321,340,1026,1299 'dash':387 'data':901 'decid':221,533,1429 'decis':275,1374 'decision-mak':274 'deep':44,1378 'default':543,610,625,645,657,1628 'defin':11,343,663,862,878,1001 'definit':338,877,1238 'deleg':146,178,188,1572 'deni':1375,1376 'descript':407,491,513,517,1109,1442,1457,1479,1564 'design':37,834 'detail':1592,1613 'direct':239,903 'directori':1366 'disabl':1203,1271,1622,1630 'discov':689 'distribut':1604 'document':47,1589,1609 'domain':51 'domain-specif':50 'e.g':1367 'edit':1191 'effect':840,934 'enabl':848,1202,1269,1285,1626 'enableag':1633,1641 'enclos':383 'encourag':1417 'enforc':334,943,1038,1256 'engin':949,1048,1341,1400 'ensur':821,919 'environ':818 'exact':863 'exampl':207,283,400,606,1011,1094,1472,1475,1503 'excess':771 'execut':653,817,956,1137,1260 'experiment':1640 'expert':1422,1436,1489 'expertis':205,1465 'explain':471 'explicit':181,238,761,981 'expos':126 'extens':1598,1600,1608,1621 'extern':898 'fair':1483 'fals':1286,1635,1642 'featur':856 'field':488,1313 'file':21,339,346,374,377,393,419,464,1022,1093,1193,1239,1362 'find':167,410,468 'fine':850,1040,1393 'fine-grain':849,1039,1392 'fix':477,480 'flash':426 'flash-preview':425 'fli':1211 'focus':80,448 'follow':250,1477 'forc':182,232 'format':375 'frontmatt':350,382,909,977 'full':1214 'gemini':3,8,32,72,189,423,607,805,1568 'gemini/agents':360 'gemini/agents/security-auditor.md':401 'generatecontentconfig':1323 'git':1125,1486,1488,1500 'github':1520 'given':1563 'global':811,994,1249 'go':278 'govern':1354 'grain':851,1041,1394 'grant':674,680,692,706,762,783,1057 'granular':944 'grep':420,1019 'group':677 'guardrail':1396 'handl':39 'happen':110 'hardcod':460 'help':531 'hire':75 'histori':123,736,741 'hyphen':510 'identifi':496 'immedi':319 'improv':841,1445 'independ':103,735 'indic':1461 'infinit':768 'inform':1388 'inherit':575,612 'inject':304,452 'inlin':587,879,1002 'instruct':195,1304,1597 'interact':106,825,923,1161,1170,1178,1511 'investig':226,286,1373 'invok':795 'isol':593,720,730,750,800,803,816,910,918,961,969,1016 'issu':1516 'job':79,440 'json':1279,1314,1639 'kind':415,539 'let':979 'letter':508 'level':359,369 'like':43,1302 'limit':1261 'list':549,873,1215 'local':416,542,1497 'localagentexecutor':296 'lookup':48 'loop':115,732,769 'lowercas':507 'main':31,57,71,121,129,140,163,192,218,271,528,615,746,1412 'maintain':916 'make':276,1504 'manag':1155,1159,1171,1188 'manual':1194 'map':287 'markdown':20,345,392,402,908,976 'match':203,1141,1317,1351 'max':430,628 'maximum':632,652 'maxtimeminut':1290 'maxturn':1288 'may':220 'mcp':561,563,565,591,691,700,702,714,880,886,1003 'mcpserver':582,998,1023 'md':347,361,371 'mean':734 'messag':1377 'min':649 'minut':656 'model':312,422,588,598,602,617,621,883,896,1258,1300,1536,1546 'modelconfig':1315,1322 'modelconfigs.overrides':1292 'move':804 'must':378,643 'my-custom-serv':1024 'my-isolated-ag':1014 'my-serv':703,717 'name':137,255,403,492,502,552,716,959,1013,1097,1140,1348 'need':1523 'node':1029 'note':307 'nudg':309 'number':509,619,630,633,650 'object':583,999 'omit':573 'one':1425 'oper':28,465,1487,1501 'optim':1408,1538 'overrid':1251,1275,1281,1316 'overridescop':1312,1318 'packag':1617 'parent':580 'part':828 'path/to/server.js':1031 'perform':229 'permiss':852,1060 'permit':1110 'persist':1167,1229,1250 'person':372 'persona':90,336 'place':353 'point':1243 'polici':942,948,1035,1047,1076,1131,1327,1340,1350,1365,1399 'policy.toml':1092 'potenti':446 'pr':1100,1107,1112,1143 'pr-creator':1099,1106,1111,1142 'prefix':1305 'prevent':767,837,932 'preview':427,609 'primari':311 'project':358 'project-level':357 'prompt':14,88,186,261,399,1416,1560 'properti':1085,1149 'protect':723,765 'protocol':590,885,1584 'provid':815,889,1240,1517 'purpos':1352 'push':1103,1115,1126 'quick':673,1201 'rather':989 're':1206 're-configur':1205 'read':418,1021 'reason':53 'recommend':1198 'recurs':722,764 'refer':1226,1401 'registri':813,995 'regress':1508 'relationship':290 'relev':1435 'reli':991 'reliabl':842,1447 'remot':545,1499,1565,1575,1587 'report':159,484 'requir':490 'research':231 'rest':1068 'restrict':97,1059,1074,1331,1382 'return':644 'rule':945,1077,1088,1096,1132,1145,1370 'run':726,1121,1277 'runconfig':1287 'ruthless':436 'safeti':1308,1326,1395 'save':117 'scenario':1473 'schema':487 'script':458 'search':421,1020,1506 'secur':405,411,437,1283,1320 'security-auditor':404,1282,1319 'see':793,1222,1397,1585,1606 'select':1534 'separ':113 'server':566,571,592,701,705,715,719,881,887,931,964,1004,1027 'session':34,581,616,1072,1180,1264,1385 'set':100,927,1301,1309,1390,1632 'settings.json':1169,1231,1247,1638 'share':362 'shell':1122 'short':516 'side':839,933 'singl':810 'site':457 'skill' 'skill-create-gemini-agent' 'slug':497 'sourc':902,1513 'source-andrader' 'special':25,91,99,198,408,1404 'special-syntax-for-subag':1403 'specialist':68,282 'specif':40,52,78,244,316,331,335,596,601,713,914,941,1034,1061,1080,1257,1272,1296,1307,1334,1559 'specifi':859 'sql':451 'standard':891 'start':379,1242 'state':845,917,936,982 'still':789 'straight':279 'string':493,514,540,599 'sub':1218 'sub-command':1217 'subag':22,23,65,66,83,93,109,124,151,154,171,175,199,227,234,245,253,317,322,328,538,667,725,738,753,774,778,781,798,801,823,906,921,940,957,972,986,1033,1044,1081,1084,1105,1138,1148,1156,1160,1189,1297,1335,1342,1358,1407,1410,1423,1478,1528,1554,1566,1576,1588,1599,1605,1611,1623,1624 'subagent-specif':939,1032 'suggest':475 'support':557 'symbol':249 'syntax':235,301,1405 'system':13,87,215,306,398,831,1303,1415 'target':1295 'task':42,148,157,202,241,1573 'team':365 'temperatur':428,618,622,1324 'think':1551 'time':654 'timeout':648 'token':118,772 'toml':951,1051,1095,1328,1361,1369 'tool':15,92,102,133,144,318,417,501,546,551,560,564,568,577,660,664,679,690,696,710,751,759,785,799,802,812,860,865,872,899,929,962,968,984,1018,1347 'toolnam':1120,1371 'toolset':62 'topic-agent' 'topic-agent-skills' 'topic-agentic-ai' 'topic-agents' 'topic-ai-agents' 'topic-cli' 'topic-installer' 'topic-manager' 'topic-skills' 'treat':1344 'trigger':1134 'tripl':386 'triple-dash':385 'tune':1526 'turn':431,629,636 'type':489 'unabl':791 'underscor':512 'unintend':838 'uniqu':495,1007 'univers':1151 'unsaf':463 'updat':974,1455 'usag':773,1221,1596 'use':170,174,197,247,264,299,314,498,556,604,613,670,870,988,1045,1162,1183,1246,1254,1266,1310,1336,1420,1453,1470,1494,1577 'user':368 'user-level':367 'via':1168 'virtual':1346 'visibl':525 'vulner':412,447,470 'want':267 'way':892,1199 'whether':1430 'wildcard':558,661,671,786 'window':105 'within':29,1619 'without':54,1065,1146,1190 'work':216 'workflow':332 'xss':454 'yaml':349,381,1012 'yes':494,515","prices":[{"id":"60f60b0f-e8a5-43e2-a3b0-312582dcdba4","listingId":"2c1f3ff0-6f18-463e-aa42-e0174f5c05bc","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"andrader","category":"jup","install_from":"skills.sh"},"createdAt":"2026-04-23T13:04:10.545Z"}],"sources":[{"listingId":"2c1f3ff0-6f18-463e-aa42-e0174f5c05bc","source":"github","sourceId":"andrader/jup/create-gemini-agent","sourceUrl":"https://github.com/andrader/jup/tree/main/skills/create-gemini-agent","isPrimary":false,"firstSeenAt":"2026-04-23T13:04:10.545Z","lastSeenAt":"2026-04-24T07:03:32.449Z"}],"details":{"listingId":"2c1f3ff0-6f18-463e-aa42-e0174f5c05bc","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"andrader","slug":"create-gemini-agent","github":{"repo":"andrader/jup","stars":10,"topics":["agent","agent-skills","agentic-ai","agents","ai","ai-agents","cli","installer","manager","skills"],"license":null,"html_url":"https://github.com/andrader/jup","pushed_at":"2026-04-24T03:07:45Z","description":"jup is a command-line tool for installing and managing agent skills across different ai agents, scopes, and projects.","skill_md_sha":"788480254e2eb8f9e34e4f51fac7051c3c1a238c","skill_md_path":"skills/create-gemini-agent/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/andrader/jup/tree/main/skills/create-gemini-agent"},"layout":"multi","source":"github","category":"jup","frontmatter":{"name":"create-gemini-agent","description":"Create a custom Gemini agent by defining its system prompt, tools, and configuration in a markdown file."},"skills_sh_url":"https://skills.sh/andrader/jup/create-gemini-agent"},"updatedAt":"2026-04-24T07:03:32.449Z"}}