[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"summary-29afff84121a00de-secure-healthcare-agents-with-bigtable-adk-model-a-summary":3,"summaries-facets-categories":107,"summary-related-29afff84121a00de-secure-healthcare-agents-with-bigtable-adk-model-a-summary":3742},{"id":4,"title":5,"ai":6,"body":13,"categories":63,"created_at":65,"date_modified":65,"description":57,"extension":66,"faq":65,"featured":67,"kicker_label":65,"meta":68,"navigation":86,"path":87,"published_at":88,"question":65,"scraped_at":89,"seo":90,"sitemap":91,"source_id":92,"source_name":93,"source_type":94,"source_url":95,"stem":96,"tags":97,"thumbnail_url":102,"tldr":103,"tweet":104,"unknown_tags":105,"__hash__":106},"summaries\u002Fsummaries\u002F29afff84121a00de-secure-healthcare-agents-with-bigtable-adk-model-a-summary.md","Secure Healthcare Agents with Bigtable, ADK & Model Armor",{"provider":7,"model":8,"input_tokens":9,"output_tokens":10,"processing_time_ms":11,"cost_usd":12},"openrouter","x-ai\u002Fgrok-4.1-fast",5383,1711,23537,0.00190945,{"type":14,"value":15,"toc":56},"minimark",[16,21,25,28,32,40,43,47,50,53],[17,18,20],"h2",{"id":19},"architect-multi-step-agents-for-personalized-healthcare-queries","Architect Multi-Step Agents for Personalized Healthcare Queries",[22,23,24],"p",{},"Root agents orchestrate sub-agents as tools: query agent accesses Bigtable via SQL for patient data (profile, prescriptions, visits, tests column families keyed by user email); search agent grounds responses with Google Search and Maps for nearby facilities; booking agent checks calendar availability and creates events. Pre-agent tools provide context—get profile info fetches age, gender, zip code for personalization; update time supplies current date for relative filters like \"past 3 months.\" Agent instructions enforce markdown tables for multi-result responses and Recharts JSON for interactive charts (e.g., test results over time with start\u002Fend dates and result limits). This agent-as-tool pattern handles operational queries efficiently, as context windows expand and token costs drop, avoiding full data dumps.",[22,26,27],{},"For complex reasoning, root agent chains sub-agent calls: retrieves medical history from Bigtable, combines with profile data, searches for age-appropriate norms (e.g., colonoscopy frequency), and synthesizes responses. Relative time parsing identifies past (March 2nd) vs. future (June 11th, 2026) dates, links descriptions (flu shot in checkup note), and associates synonyms (high blood pressure = hypertension → cardiologist).",[17,29,31],{"id":30},"enforce-row-level-security-and-type-safety-in-database-tools","Enforce Row-Level Security and Type Safety in Database Tools",[22,33,34,35,39],{},"Bigtable's ADK query tool converts SQL into function tools in few lines: parameterize queries with Pydantic for type enforcement (dates, integers) blocking SQL injection (e.g., attempts to bypass row key filters). Tool context passes user identity (from OAuth scopes including Google Calendar) for row-level access control (RLAC)—filter ",[36,37,38],"code",{},"WHERE row_key = user_email"," without agent visibility, preventing impersonation since agents run as service accounts.",[22,41,42],{},"Column families act as SQL maps: tests family keys test types (glucose, cholesterol) with timestamped JSON values. Single queries suffice for most user data (recent orders\u002Ftickets); for heavy analytics, use Data Agent Developer Platform (faceted search) or Conversational Analytics API.",[17,44,46],{"id":45},"iterate-with-evals-self-tuning-and-injection-guards","Iterate with Evals, Self-Tuning, and Injection Guards",[22,48,49],{},"ADK evals visualize agent traces (sub-agent calls, tool inputs) to debug regressions during instruction\u002Ftool tweaks. Implement hill-climbing loops: agent runs evals, suggests improvements (e.g., better prompts), applies changes, re-runs—achieving self-tuning without manual intervention.",[22,51,52],{},"Integrate ADK calendar tools (find availability, create\u002Fmodify\u002Fcancel events) scoped to user identity. Guard inputs\u002Foutputs with Model Armor templates (injection, jailbreaking controls) via callbacks: blocks prompt injections like \"ignore previous instructions\" and sensitive data (SSNs, credit cards). OAuth setup requires client ID\u002Fsecret, origins\u002Fredirect URIs, and scopes for calendar\u002Fdatabase access.",[22,54,55],{},"This stack delivers production-ready agents: authenticates users, personalizes via DB+context, reasons multi-step, visualizes data, books actions, and stays secure—demo data randomly generated in Bigtable patients table.",{"title":57,"searchDepth":58,"depth":58,"links":59},"",2,[60,61,62],{"id":19,"depth":58,"text":20},{"id":30,"depth":58,"text":31},{"id":45,"depth":58,"text":46},[64],"AI Automation",null,"md",false,{"content_references":69,"triage":81},[70,75,77,79],{"type":71,"title":72,"url":73,"context":74},"tool","Bigtable","https:\u002F\u002Fgoo.gle\u002F4wqWan1","mentioned",{"type":71,"title":76,"context":74},"Agent Development Kit (ADK)",{"type":71,"title":78,"context":74},"Model Armor",{"type":71,"title":80,"context":74},"Recharts",{"relevance":82,"novelty":83,"quality":82,"actionability":82,"composite":84,"reasoning":85},4,3,3.8,"Category: AI & LLMs. The article discusses building personalized healthcare agents using AI tools, addressing practical applications of AI in healthcare, which is relevant to the target audience. It provides specific examples of agent architecture and security measures, making it actionable for developers looking to implement similar solutions.",true,"\u002Fsummaries\u002F29afff84121a00de-secure-healthcare-agents-with-bigtable-adk-model-a-summary","2026-05-14 22:13:53","2026-05-14 23:00:28",{"title":5,"description":57},{"loc":87},"29afff84121a00de","Google Cloud Tech","video","https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=HE1ldmzennU","summaries\u002F29afff84121a00de-secure-healthcare-agents-with-bigtable-adk-model-a-summary",[98,99,100,101],"agents","ai-tools","cloud","devops-cloud","https:\u002F\u002Fi.ytimg.com\u002Fvi\u002FHE1ldmzennU\u002Fhqdefault.jpg","Build personalized conversational agents using Bigtable's SQL query tools via ADK for secure user data access, sub-agents for multi-step reasoning, calendar integration for bookings, and Model Armor to block SQL\u002Fprompt injections.","Demo of a healthcare agent built with [Bigtable](https:\u002F\u002Fgoo.gle\u002F4wqWan1) for patient data (via SQL API with row-level security), ADK for query tools\u002Fsub-agents\u002Fcalendar integration\u002Fevals, and Model Armor for SQL\u002Fprompt injection guards. Walks through auth flow, multi-step reasoning, and self-tuning evals.",[101],"_6t15eUlWp5BsueQIcUrU30nl-ZaoY8QybEFHLLFRdY",[108,111,114,117,119,122,124,126,128,130,132,134,137,139,141,143,145,147,149,151,153,155,158,161,163,165,168,170,172,175,177,179,181,183,185,187,189,191,193,195,197,199,201,203,205,207,209,211,213,215,217,219,221,223,225,227,229,231,233,235,237,239,241,243,245,247,249,251,253,255,257,259,261,263,265,267,269,271,274,276,278,280,282,284,286,288,290,292,294,296,298,300,302,304,306,308,310,312,314,316,318,320,322,324,326,328,330,332,334,336,338,340,342,344,346,348,350,352,354,356,358,360,362,364,366,368,370,372,374,376,378,380,382,384,386,388,390,392,394,396,398,400,402,404,406,408,410,412,414,416,418,420,422,424,426,428,430,432,434,436,438,440,442,444,446,448,450,452,454,456,458,460,462,464,466,468,470,472,474,476,478,480,482,484,486,488,490,492,494,496,498,500,502,504,506,508,510,512,514,516,518,520,522,524,526,528,530,532,534,536,538,540,542,544,546,548,550,552,554,556,558,560,562,564,566,568,570,572,574,576,578,580,582,584,586,588,590,592,594,596,598,600,602,604,606,608,610,612,614,616,618,620,622,624,626,628,630,632,634,636,638,640,642,644,646,648,650,652,654,656,658,660,662,664,666,668,670,672,674,676,678,680,682,684,686,688,690,692,694,696,698,700,702,704,706,708,710,712,714,716,718,720,722,724,726,728,730,732,734,736,738,740,742,744,746,748,750,752,754,756,758,760,762,764,766,768,770,772,774,776,778,780,782,784,786,788,790,792,794,796,798,800,802,804,806,808,810,812,814,816,818,820,822,824,826,828,830,832,834,836,838,840,842,844,846,848,850,852,854,856,858,860,862,864,866,868,870,872,874,876,878,880,882,884,886,888,890,892,894,896,898,900,902,904,906,908,910,912,914,916,918,920,922,924,926,928,930,932,934,936,938,940,942,944,946,948,950,952,954,956,958,960,962,964,966,968,970,972,974,976,978,980,982,984,986,988,990,992,994,996,998,1000,1002,1004,1006,1008,1010,1012,1014,1016,1018,1020,1022,1024,1026,1028,1030,1032,1034,1036,1038,1040,1042,1044,1046,1048,1050,1052,1054,1056,1058,1060,1062,1064,1066,1068,1070,1072,1074,1076,1078,1080,1082,1084,1086,1088,1090,1092,1094,1096,1098,1100,1102,1104,1106,1108,1110,1112,1114,1116,1118,1120,1122,1124,1126,1128,1130,1132,1134,1136,1138,1140,1142,1144,1146,1148,1150,1152,1154,1156,1158,1160,1162,1164,1166,1168,1170,1172,1174,1176,1178,1180,1182,1184,1186,1188,1190,1192,1194,1196,1198,1200,1202,1204,1206,1208,1210,1212,1214,1216,1218,1220,1222,1224,1226,1228,1230,1232,1234,1236,1238,1240,1242,1244,1246,1248,1250,1252,1254,1256,1258,1260,1262,1264,1266,1268,1270,1272,1274,1276,1278,1280,1282,1284,1286,1288,1290,1292,1294,1296,1298,1300,1302,1304,1306,1308,1310,1312,1314,1316,1318,1320,1322,1324,1326,1328,1330,1332,1334,1336,1338,1340,1342,1344,1346,1348,1350,1352,1354,1356,1358,1360,1362,1364,1366,1368,1370,1372,1374,1376,1378,1380,1382,1384,1386,1388,1390,1392,1394,1396,1398,1400,1402,1404,1406,1408,1410,1412,1414,1416,1418,1420,1422,1424,1426,1428,1430,1432,1434,1436,1438,1440,1442,1444,1446,1448,1450,1452,1454,1456,1458,1460,1462,1464,1466,1468,1470,1472,1474,1476,1478,1480,1482,1484,1486,1488,1490,1492,1494,1496,1498,1500,1502,1504,1506,1508,1510,1512,1514,1516,1518,1520,1522,1524,1526,1528,1530,1532,1534,1536,1538,1540,1542,1544,1546,1548,1550,1552,1554,1556,1558,1560,1562,1564,1566,1568,1570,1572,1574,1576,1578,1580,1582,1584,1586,1588,1590,1592,1594,1596,1598,1600,1602,1604,1606,1608,1610,1612,1614,1616,1618,1620,1622,1624,1626,1628,1630,1632,1634,1636,1638,1640,1642,1644,1646,1648,1650,1652,1654,1656,1658,1660,1662,1664,1666,1668,1670,1672,1674,1676,1678,1680,1682,1684,1686,1688,1690,1692,1694,1696,1698,1700,1702,1704,1706,1708,1710,1712,1714,1716,1718,1720,1722,1724,1726,1728,1730,1732,1734,1736,1738,1740,1742,1744,1746,1748,1750,1752,1754,1756,1758,1760,1762,1764,1766,1768,1770,1772,1774,1776,1778,1780,1782,1784,1786,1788,1790,1792,1794,1796,1798,1800,1802,1804,1806,1808,1810,1812,1814,1816,1818,1820,1822,1824,1826,1828,1830,1832,1834,1836,1838,1840,1842,1844,1846,1848,1850,1852,1854,1856,1858,1860,1862,1864,1866,1868,1870,1872,1874,1876,1878,1880,1882,1884,1886,1888,1890,1892,1894,1896,1898,1900,1902,1904,1906,1908,1910,1912,1914,1916,1918,1920,1922,1924,1926,1928,1930,1932,1934,1936,1938,1940,1942,1944,1946,1948,1950,1952,1954,1956,1958,1960,1962,1964,1966,1968,1970,1972,1974,1976,1978,1980,1982,1984,1986,1988,1990,1992,1994,1996,1998,2000,2002,2004,2006,2008,2010,2012,2014,2016,2018,2020,2022,2024,2026,2028,2030,2032,2034,2036,2038,2040,2042,2044,2046,2048,2050,2052,2054,2056,2058,2060,2062,2064,2066,2068,2070,2072,2074,2076,2078,2080,2082,2084,2086,2088,2090,2092,2094,2096,2098,2100,2102,2104,2106,2108,2110,2112,2114,2116,2118,2120,2122,2124,2126,2128,2130,2132,2134,2136,2138,2140,2142,2144,2146,2148,2150,2152,2154,2156,2158,2160,2162,2164,2166,2168,2170,2172,2174,2176,2178,2180,2182,2184,2186,2188,2190,2192,2194,2196,2198,2200,2202,2204,2206,2208,2210,2212,2214,2216,2218,2220,2222,2224,2226,2228,2230,2232,2234,2236,2238,2240,2242,2244,2246,2248,2250,2252,2254,2256,2258,2260,2262,2264,2266,2268,2270,2272,2274,2276,2278,2280,2282,2284,2286,2288,2290,2292,2294,2296,2298,2300,2302,2304,2306,2308,2310,2312,2314,2316,2318,2320,2322,2324,2326,2328,2330,2332,2334,2336,2338,2340,2342,2344,2346,2348,2350,2352,2354,2356,2358,2360,2362,2364,2366,2368,2370,2372,2374,2376,2378,2380,2382,2384,2386,2388,2390,2392,2394,2396,2398,2400,2402,2404,2406,2408,2410,2412,2414,2416,2418,2420,2422,2424,2426,2428,2430,2432,2434,2436,2438,2440,2442,2444,2446,2448,2450,2452,2454,2456,2458,2460,2462,2464,2466,2468,2470,2472,2474,2476,2478,2480,2482,2484,2486,2488,2490,2492,2494,2496,2498,2500,2502,2504,2506,2508,2510,2512,2514,2516,2518,2520,2522,2524,2526,2528,2530,2532,2534,2536,2538,2540,2542,2544,2546,2548,2550,2552,2554,2556,2558,2560,2562,2564,2566,2568,2570,2572,2574,2576,2578,2580,2582,2584,2586,2588,2590,2592,2594,2596,2598,2600,2602,2604,2606,2608,2610,2612,2614,2616,2618,2620,2622,2624,2626,2628,2630,2632,2634,2636,2638,2640,2642,2644,2646,2648,2650,2652,2654,2656,2658,2660,2662,2664,2666,2668,2670,2672,2674,2676,2678,2680,2682,2684,2686,2688,2690,2692,2694,2696,2698,2700,2702,2704,2706,2708,2710,2712,2714,2716,2718,2720,2722,2724,2726,2728,2730,2732,2734,2736,2738,2740,2742,2744,2746,2748,2750,2752,2754,2756,2758,2760,2762,2764,2766,2768,2770,2772,2774,2776,2778,2780,2782,2784,2786,2788,2790,2792,2794,2796,2798,2800,2802,2804,2806,2808,2810,2812,2814,2816,2818,2820,2822,2824,2826,2828,2830,2832,2834,2836,2838,2840,2842,2844,2846,2848,2850,2852,2854,2856,2858,2860,2862,2864,2866,2868,2870,2872,2874,2876,2878,2880,2882,2884,2886,2888,2890,2892,2894,2896,2898,2900,2902,2904,2906,2908,2910,2912,2914,2916,2918,2920,2922,2924,2926,2928,2930,2932,2934,2936,2938,2940,2942,2944,2946,2948,2950,2952,2954,2956,2958,2960,2962,2964,2966,2968,2970,2972,2974,2976,2978,2980,2982,2984,2986,2988,2990,2992,2994,2996,2998,3000,3002,3004,3006,3008,3010,3012,3014,3016,3018,3020,3022,3024,3026,3028,3030,3032,3034,3036,3038,3040,3042,3044,3046,3048,3050,3052,3054,3056,3058,3060,3062,3064,3066,3068,3070,3072,3074,3076,3078,3080,3082,3084,3086,3088,3090,3092,3094,3096,3098,3100,3102,3104,3106,3108,3110,3112,3114,3116,3118,3120,3122,3124,3126,3128,3130,3132,3134,3136,3138,3140,3142,3144,3146,3148,3150,3152,3154,3156,3158,3160,3162,3164,3166,3168,3170,3172,3174,3176,3178,3180,3182,3184,3186,3188,3190,3192,3194,3196,3198,3200,3202,3204,3206,3208,3210,3212,3214,3216,3218,3220,3222,3224,3226,3228,3230,3232,3234,3236,3238,3240,3242,3244,3246,3248,3250,3252,3254,3256,3258,3260,3262,3264,3266,3268,3270,3272,3274,3276,3278,3280,3282,3284,3286,3288,3290,3292,3294,3296,3298,3300,3302,3304,3306,3308,3310,3312,3314,3316,3318,3320,3322,3324,3326,3328,3330,3332,3334,3336,3338,3340,3342,3344,3346,3348,3350,3352,3354,3356,3358,3360,3362,3364,3366,3368,3370,3372,3374,3376,3378,3380,3382,3384,3386,3388,3390,3392,3394,3396,3398,3400,3402,3404,3406,3408,3410,3412,3414,3416,3418,3420,3422,3424,3426,3428,3430,3432,3434,3436,3438,3440,3442,3444,3446,3448,3450,3452,3454,3456,3458,3460,3462,3464,3466,3468,3470,3472,3474,3476,3478,3480,3482,3484,3486,3488,3490,3492,3494,3496,3498,3500,3502,3504,3506,3508,3510,3512,3514,3516,3518,3520,3522,3524,3526,3528,3530,3532,3534,3536,3538,3540,3542,3544,3546,3548,3550,3552,3554,3556,3558,3560,3562,3564,3566,3568,3570,3572,3574,3576,3578,3580,3582,3584,3586,3588,3590,3592,3594,3596,3598,3600,3602,3604,3606,3608,3610,3612,3614,3616,3618,3620,3622,3624,3626,3628,3630,3632,3634,3636,3638,3640,3642,3644,3646,3648,3650,3652,3654,3656,3658,3660,3662,3664,3666,3668,3670,3672,3674,3676,3678,3680,3682,3684,3686,3688,3690,3692,3694,3696,3698,3700,3702,3704,3706,3708,3710,3712,3714,3716,3718,3720,3722,3724,3726,3728,3730,3732,3734,3736,3738,3740],{"categories":109},[110],"Developer Productivity",{"categories":112},[113],"Business & SaaS",{"categories":115},[116],"AI & LLMs",{"categories":118},[64],{"categories":120},[121],"Product Strategy",{"categories":123},[116],{"categories":125},[110],{"categories":127},[113],{"categories":129},[],{"categories":131},[116],{"categories":133},[],{"categories":135},[136],"AI News & Trends",{"categories":138},[64],{"categories":140},[136],{"categories":142},[64],{"categories":144},[64],{"categories":146},[116],{"categories":148},[116],{"categories":150},[136],{"categories":152},[116],{"categories":154},[],{"categories":156},[157],"Design & Frontend",{"categories":159},[160],"Data Science & Visualization",{"categories":162},[136],{"categories":164},[],{"categories":166},[167],"Software Engineering",{"categories":169},[116],{"categories":171},[64],{"categories":173},[174],"Marketing & Growth",{"categories":176},[116],{"categories":178},[64],{"categories":180},[],{"categories":182},[],{"categories":184},[157],{"categories":186},[64],{"categories":188},[110],{"categories":190},[157],{"categories":192},[116],{"categories":194},[64],{"categories":196},[136],{"categories":198},[],{"categories":200},[],{"categories":202},[64],{"categories":204},[167],{"categories":206},[],{"categories":208},[113],{"categories":210},[],{"categories":212},[],{"categories":214},[64],{"categories":216},[64],{"categories":218},[116],{"categories":220},[],{"categories":222},[167],{"categories":224},[],{"categories":226},[],{"categories":228},[],{"categories":230},[116],{"categories":232},[174],{"categories":234},[157],{"categories":236},[157],{"categories":238},[116],{"categories":240},[64],{"categories":242},[116],{"categories":244},[116],{"categories":246},[64],{"categories":248},[64],{"categories":250},[160],{"categories":252},[136],{"categories":254},[64],{"categories":256},[174],{"categories":258},[64],{"categories":260},[121],{"categories":262},[],{"categories":264},[64],{"categories":266},[],{"categories":268},[64],{"categories":270},[167],{"categories":272},[273],"DevOps & Cloud",{"categories":275},[157],{"categories":277},[116],{"categories":279},[],{"categories":281},[],{"categories":283},[64],{"categories":285},[],{"categories":287},[116],{"categories":289},[],{"categories":291},[110],{"categories":293},[167],{"categories":295},[113],{"categories":297},[136],{"categories":299},[116],{"categories":301},[],{"categories":303},[116],{"categories":305},[],{"categories":307},[167],{"categories":309},[160],{"categories":311},[],{"categories":313},[116],{"categories":315},[157],{"categories":317},[],{"categories":319},[157],{"categories":321},[64],{"categories":323},[],{"categories":325},[64],{"categories":327},[136],{"categories":329},[113],{"categories":331},[116],{"categories":333},[],{"categories":335},[64],{"categories":337},[116],{"categories":339},[121],{"categories":341},[],{"categories":343},[116],{"categories":345},[64],{"categories":347},[64],{"categories":349},[],{"categories":351},[160],{"categories":353},[116],{"categories":355},[],{"categories":357},[110],{"categories":359},[113],{"categories":361},[116],{"categories":363},[64],{"categories":365},[167],{"categories":367},[116],{"categories":369},[],{"categories":371},[],{"categories":373},[116],{"categories":375},[],{"categories":377},[157],{"categories":379},[],{"categories":381},[116],{"categories":383},[],{"categories":385},[64],{"categories":387},[116],{"categories":389},[157],{"categories":391},[],{"categories":393},[116],{"categories":395},[116],{"categories":397},[113],{"categories":399},[64],{"categories":401},[116],{"categories":403},[157],{"categories":405},[64],{"categories":407},[],{"categories":409},[],{"categories":411},[136],{"categories":413},[],{"categories":415},[116],{"categories":417},[113,174],{"categories":419},[],{"categories":421},[116],{"categories":423},[64],{"categories":425},[],{"categories":427},[],{"categories":429},[116],{"categories":431},[],{"categories":433},[116],{"categories":435},[273],{"categories":437},[],{"categories":439},[136],{"categories":441},[157],{"categories":443},[],{"categories":445},[136],{"categories":447},[136],{"categories":449},[116],{"categories":451},[174],{"categories":453},[],{"categories":455},[113],{"categories":457},[],{"categories":459},[116,273],{"categories":461},[116],{"categories":463},[116],{"categories":465},[116],{"categories":467},[64],{"categories":469},[116,167],{"categories":471},[160],{"categories":473},[116],{"categories":475},[174],{"categories":477},[64],{"categories":479},[64],{"categories":481},[],{"categories":483},[64],{"categories":485},[116],{"categories":487},[116,113],{"categories":489},[],{"categories":491},[157],{"categories":493},[157],{"categories":495},[],{"categories":497},[],{"categories":499},[136],{"categories":501},[],{"categories":503},[110],{"categories":505},[167],{"categories":507},[116],{"categories":509},[157],{"categories":511},[64],{"categories":513},[167],{"categories":515},[136],{"categories":517},[157],{"categories":519},[],{"categories":521},[116],{"categories":523},[116],{"categories":525},[116],{"categories":527},[136],{"categories":529},[110],{"categories":531},[116],{"categories":533},[64],{"categories":535},[273],{"categories":537},[157],{"categories":539},[64],{"categories":541},[],{"categories":543},[],{"categories":545},[157],{"categories":547},[136],{"categories":549},[160],{"categories":551},[],{"categories":553},[116],{"categories":555},[116],{"categories":557},[113],{"categories":559},[116],{"categories":561},[116],{"categories":563},[136],{"categories":565},[],{"categories":567},[64],{"categories":569},[167],{"categories":571},[],{"categories":573},[116],{"categories":575},[116],{"categories":577},[64],{"categories":579},[],{"categories":581},[],{"categories":583},[116],{"categories":585},[],{"categories":587},[113],{"categories":589},[64],{"categories":591},[],{"categories":593},[110],{"categories":595},[116],{"categories":597},[113],{"categories":599},[136],{"categories":601},[110],{"categories":603},[],{"categories":605},[],{"categories":607},[],{"categories":609},[136],{"categories":611},[136],{"categories":613},[],{"categories":615},[],{"categories":617},[113],{"categories":619},[],{"categories":621},[],{"categories":623},[110],{"categories":625},[],{"categories":627},[174],{"categories":629},[64],{"categories":631},[113],{"categories":633},[64],{"categories":635},[167],{"categories":637},[],{"categories":639},[121],{"categories":641},[157],{"categories":643},[167],{"categories":645},[116],{"categories":647},[64],{"categories":649},[113],{"categories":651},[116],{"categories":653},[],{"categories":655},[],{"categories":657},[167],{"categories":659},[160],{"categories":661},[121],{"categories":663},[64],{"categories":665},[116],{"categories":667},[],{"categories":669},[273],{"categories":671},[],{"categories":673},[64],{"categories":675},[],{"categories":677},[110],{"categories":679},[],{"categories":681},[116],{"categories":683},[157],{"categories":685},[174],{"categories":687},[64],{"categories":689},[],{"categories":691},[110],{"categories":693},[],{"categories":695},[136],{"categories":697},[116,273],{"categories":699},[136],{"categories":701},[116],{"categories":703},[113],{"categories":705},[116],{"categories":707},[],{"categories":709},[113],{"categories":711},[],{"categories":713},[167],{"categories":715},[157],{"categories":717},[136],{"categories":719},[160],{"categories":721},[110],{"categories":723},[116],{"categories":725},[64],{"categories":727},[167],{"categories":729},[],{"categories":731},[],{"categories":733},[121],{"categories":735},[],{"categories":737},[116],{"categories":739},[],{"categories":741},[157],{"categories":743},[157],{"categories":745},[157],{"categories":747},[],{"categories":749},[],{"categories":751},[136],{"categories":753},[64],{"categories":755},[116],{"categories":757},[116],{"categories":759},[116],{"categories":761},[113],{"categories":763},[116],{"categories":765},[],{"categories":767},[167],{"categories":769},[167],{"categories":771},[113],{"categories":773},[],{"categories":775},[116],{"categories":777},[116],{"categories":779},[113],{"categories":781},[136],{"categories":783},[174],{"categories":785},[64],{"categories":787},[],{"categories":789},[157],{"categories":791},[],{"categories":793},[116],{"categories":795},[],{"categories":797},[113],{"categories":799},[64],{"categories":801},[],{"categories":803},[273],{"categories":805},[160],{"categories":807},[167],{"categories":809},[174],{"categories":811},[167],{"categories":813},[64],{"categories":815},[],{"categories":817},[],{"categories":819},[64],{"categories":821},[110],{"categories":823},[64],{"categories":825},[121],{"categories":827},[113],{"categories":829},[],{"categories":831},[116],{"categories":833},[121],{"categories":835},[116],{"categories":837},[116],{"categories":839},[174],{"categories":841},[157],{"categories":843},[64],{"categories":845},[],{"categories":847},[],{"categories":849},[273],{"categories":851},[167],{"categories":853},[],{"categories":855},[64],{"categories":857},[116],{"categories":859},[157,116],{"categories":861},[110],{"categories":863},[],{"categories":865},[116],{"categories":867},[110],{"categories":869},[157],{"categories":871},[64],{"categories":873},[167],{"categories":875},[],{"categories":877},[116],{"categories":879},[],{"categories":881},[],{"categories":883},[110],{"categories":885},[],{"categories":887},[64],{"categories":889},[121],{"categories":891},[116],{"categories":893},[116],{"categories":895},[157],{"categories":897},[64],{"categories":899},[273],{"categories":901},[157],{"categories":903},[64],{"categories":905},[116],{"categories":907},[116],{"categories":909},[116],{"categories":911},[167],{"categories":913},[],{"categories":915},[136],{"categories":917},[],{"categories":919},[121],{"categories":921},[64],{"categories":923},[157],{"categories":925},[64],{"categories":927},[167],{"categories":929},[157],{"categories":931},[64],{"categories":933},[136],{"categories":935},[],{"categories":937},[116],{"categories":939},[157],{"categories":941},[116],{"categories":943},[110],{"categories":945},[136],{"categories":947},[116],{"categories":949},[174],{"categories":951},[116],{"categories":953},[116],{"categories":955},[64],{"categories":957},[64],{"categories":959},[116],{"categories":961},[64],{"categories":963},[157],{"categories":965},[116],{"categories":967},[],{"categories":969},[],{"categories":971},[167],{"categories":973},[],{"categories":975},[110],{"categories":977},[273],{"categories":979},[],{"categories":981},[110],{"categories":983},[113],{"categories":985},[174],{"categories":987},[],{"categories":989},[113],{"categories":991},[],{"categories":993},[],{"categories":995},[],{"categories":997},[],{"categories":999},[],{"categories":1001},[116],{"categories":1003},[64],{"categories":1005},[273],{"categories":1007},[110],{"categories":1009},[116],{"categories":1011},[167],{"categories":1013},[121],{"categories":1015},[116],{"categories":1017},[174],{"categories":1019},[116],{"categories":1021},[116],{"categories":1023},[116],{"categories":1025},[116,110],{"categories":1027},[167],{"categories":1029},[167],{"categories":1031},[157],{"categories":1033},[116],{"categories":1035},[],{"categories":1037},[],{"categories":1039},[],{"categories":1041},[167],{"categories":1043},[160],{"categories":1045},[136],{"categories":1047},[157],{"categories":1049},[],{"categories":1051},[116],{"categories":1053},[116],{"categories":1055},[],{"categories":1057},[],{"categories":1059},[64],{"categories":1061},[116],{"categories":1063},[113],{"categories":1065},[],{"categories":1067},[110],{"categories":1069},[116],{"categories":1071},[110],{"categories":1073},[116],{"categories":1075},[167],{"categories":1077},[174],{"categories":1079},[116,157],{"categories":1081},[136],{"categories":1083},[157],{"categories":1085},[],{"categories":1087},[273],{"categories":1089},[157],{"categories":1091},[64],{"categories":1093},[],{"categories":1095},[],{"categories":1097},[],{"categories":1099},[],{"categories":1101},[167],{"categories":1103},[64],{"categories":1105},[64],{"categories":1107},[273],{"categories":1109},[116],{"categories":1111},[116],{"categories":1113},[116],{"categories":1115},[],{"categories":1117},[157],{"categories":1119},[],{"categories":1121},[],{"categories":1123},[64],{"categories":1125},[],{"categories":1127},[],{"categories":1129},[174],{"categories":1131},[174],{"categories":1133},[64],{"categories":1135},[],{"categories":1137},[116],{"categories":1139},[116],{"categories":1141},[167],{"categories":1143},[157],{"categories":1145},[157],{"categories":1147},[64],{"categories":1149},[110],{"categories":1151},[116],{"categories":1153},[157],{"categories":1155},[157],{"categories":1157},[64],{"categories":1159},[64],{"categories":1161},[116],{"categories":1163},[],{"categories":1165},[],{"categories":1167},[116],{"categories":1169},[64],{"categories":1171},[136],{"categories":1173},[167],{"categories":1175},[110],{"categories":1177},[116],{"categories":1179},[],{"categories":1181},[64],{"categories":1183},[64],{"categories":1185},[],{"categories":1187},[110],{"categories":1189},[116],{"categories":1191},[110],{"categories":1193},[110],{"categories":1195},[],{"categories":1197},[],{"categories":1199},[64],{"categories":1201},[64],{"categories":1203},[116],{"categories":1205},[116],{"categories":1207},[136],{"categories":1209},[160],{"categories":1211},[121],{"categories":1213},[136],{"categories":1215},[157],{"categories":1217},[],{"categories":1219},[],{"categories":1221},[136],{"categories":1223},[],{"categories":1225},[],{"categories":1227},[],{"categories":1229},[],{"categories":1231},[167],{"categories":1233},[160],{"categories":1235},[],{"categories":1237},[116],{"categories":1239},[116],{"categories":1241},[160],{"categories":1243},[167],{"categories":1245},[],{"categories":1247},[],{"categories":1249},[64],{"categories":1251},[136],{"categories":1253},[136],{"categories":1255},[64],{"categories":1257},[110],{"categories":1259},[116,273],{"categories":1261},[],{"categories":1263},[157],{"categories":1265},[110],{"categories":1267},[64],{"categories":1269},[157],{"categories":1271},[],{"categories":1273},[64],{"categories":1275},[64],{"categories":1277},[116],{"categories":1279},[174],{"categories":1281},[167],{"categories":1283},[157],{"categories":1285},[],{"categories":1287},[64],{"categories":1289},[116],{"categories":1291},[64],{"categories":1293},[64],{"categories":1295},[64],{"categories":1297},[174],{"categories":1299},[64],{"categories":1301},[116],{"categories":1303},[],{"categories":1305},[174],{"categories":1307},[136],{"categories":1309},[64],{"categories":1311},[],{"categories":1313},[],{"categories":1315},[116],{"categories":1317},[64],{"categories":1319},[136],{"categories":1321},[64],{"categories":1323},[],{"categories":1325},[],{"categories":1327},[],{"categories":1329},[64],{"categories":1331},[],{"categories":1333},[],{"categories":1335},[160],{"categories":1337},[116],{"categories":1339},[160],{"categories":1341},[136],{"categories":1343},[116],{"categories":1345},[116],{"categories":1347},[64],{"categories":1349},[116],{"categories":1351},[],{"categories":1353},[],{"categories":1355},[273],{"categories":1357},[],{"categories":1359},[],{"categories":1361},[110],{"categories":1363},[],{"categories":1365},[],{"categories":1367},[],{"categories":1369},[],{"categories":1371},[167],{"categories":1373},[136],{"categories":1375},[174],{"categories":1377},[113],{"categories":1379},[116],{"categories":1381},[116],{"categories":1383},[113],{"categories":1385},[],{"categories":1387},[157],{"categories":1389},[64],{"categories":1391},[113],{"categories":1393},[116],{"categories":1395},[116],{"categories":1397},[110],{"categories":1399},[],{"categories":1401},[110],{"categories":1403},[116],{"categories":1405},[174],{"categories":1407},[64],{"categories":1409},[136],{"categories":1411},[113],{"categories":1413},[116],{"categories":1415},[64],{"categories":1417},[],{"categories":1419},[116],{"categories":1421},[110],{"categories":1423},[116],{"categories":1425},[],{"categories":1427},[136],{"categories":1429},[116],{"categories":1431},[],{"categories":1433},[113],{"categories":1435},[113],{"categories":1437},[116],{"categories":1439},[],{"categories":1441},[],{"categories":1443},[],{"categories":1445},[116],{"categories":1447},[],{"categories":1449},[273],{"categories":1451},[116],{"categories":1453},[],{"categories":1455},[116],{"categories":1457},[116],{"categories":1459},[116],{"categories":1461},[116,273],{"categories":1463},[116],{"categories":1465},[116],{"categories":1467},[157],{"categories":1469},[64],{"categories":1471},[],{"categories":1473},[64],{"categories":1475},[116],{"categories":1477},[116],{"categories":1479},[116],{"categories":1481},[110],{"categories":1483},[110],{"categories":1485},[167],{"categories":1487},[157],{"categories":1489},[64],{"categories":1491},[],{"categories":1493},[116],{"categories":1495},[136],{"categories":1497},[116],{"categories":1499},[113],{"categories":1501},[],{"categories":1503},[273],{"categories":1505},[157],{"categories":1507},[157],{"categories":1509},[64],{"categories":1511},[136],{"categories":1513},[64],{"categories":1515},[116],{"categories":1517},[],{"categories":1519},[116],{"categories":1521},[],{"categories":1523},[],{"categories":1525},[116],{"categories":1527},[116],{"categories":1529},[116],{"categories":1531},[64],{"categories":1533},[116],{"categories":1535},[],{"categories":1537},[160],{"categories":1539},[64],{"categories":1541},[],{"categories":1543},[],{"categories":1545},[116],{"categories":1547},[136],{"categories":1549},[],{"categories":1551},[157],{"categories":1553},[273],{"categories":1555},[136],{"categories":1557},[167],{"categories":1559},[167],{"categories":1561},[136],{"categories":1563},[136],{"categories":1565},[273],{"categories":1567},[],{"categories":1569},[136],{"categories":1571},[116],{"categories":1573},[110],{"categories":1575},[136],{"categories":1577},[],{"categories":1579},[160],{"categories":1581},[136],{"categories":1583},[167],{"categories":1585},[136],{"categories":1587},[273],{"categories":1589},[116],{"categories":1591},[116],{"categories":1593},[],{"categories":1595},[113],{"categories":1597},[],{"categories":1599},[],{"categories":1601},[116],{"categories":1603},[116],{"categories":1605},[116],{"categories":1607},[116],{"categories":1609},[],{"categories":1611},[160],{"categories":1613},[110],{"categories":1615},[],{"categories":1617},[116],{"categories":1619},[116],{"categories":1621},[273],{"categories":1623},[273],{"categories":1625},[],{"categories":1627},[64],{"categories":1629},[136],{"categories":1631},[136],{"categories":1633},[116],{"categories":1635},[64],{"categories":1637},[],{"categories":1639},[157],{"categories":1641},[116],{"categories":1643},[116],{"categories":1645},[],{"categories":1647},[],{"categories":1649},[273],{"categories":1651},[116],{"categories":1653},[167],{"categories":1655},[113],{"categories":1657},[116],{"categories":1659},[],{"categories":1661},[64],{"categories":1663},[110],{"categories":1665},[110],{"categories":1667},[],{"categories":1669},[116],{"categories":1671},[157],{"categories":1673},[64],{"categories":1675},[],{"categories":1677},[116],{"categories":1679},[116],{"categories":1681},[64],{"categories":1683},[],{"categories":1685},[64],{"categories":1687},[167],{"categories":1689},[],{"categories":1691},[116],{"categories":1693},[],{"categories":1695},[116],{"categories":1697},[],{"categories":1699},[116],{"categories":1701},[116],{"categories":1703},[],{"categories":1705},[116],{"categories":1707},[136],{"categories":1709},[116],{"categories":1711},[116],{"categories":1713},[110],{"categories":1715},[116],{"categories":1717},[136],{"categories":1719},[64],{"categories":1721},[],{"categories":1723},[116],{"categories":1725},[174],{"categories":1727},[],{"categories":1729},[],{"categories":1731},[],{"categories":1733},[110],{"categories":1735},[136],{"categories":1737},[64],{"categories":1739},[116],{"categories":1741},[157],{"categories":1743},[64],{"categories":1745},[],{"categories":1747},[64],{"categories":1749},[],{"categories":1751},[116],{"categories":1753},[64],{"categories":1755},[116],{"categories":1757},[],{"categories":1759},[116],{"categories":1761},[116],{"categories":1763},[136],{"categories":1765},[157],{"categories":1767},[64],{"categories":1769},[157],{"categories":1771},[113],{"categories":1773},[],{"categories":1775},[],{"categories":1777},[116],{"categories":1779},[110],{"categories":1781},[136],{"categories":1783},[],{"categories":1785},[],{"categories":1787},[167],{"categories":1789},[157],{"categories":1791},[],{"categories":1793},[116],{"categories":1795},[],{"categories":1797},[174],{"categories":1799},[116],{"categories":1801},[273],{"categories":1803},[167],{"categories":1805},[],{"categories":1807},[64],{"categories":1809},[116],{"categories":1811},[64],{"categories":1813},[64],{"categories":1815},[116],{"categories":1817},[],{"categories":1819},[110],{"categories":1821},[116],{"categories":1823},[113],{"categories":1825},[167],{"categories":1827},[157],{"categories":1829},[],{"categories":1831},[],{"categories":1833},[],{"categories":1835},[64],{"categories":1837},[157],{"categories":1839},[136],{"categories":1841},[116],{"categories":1843},[136],{"categories":1845},[157],{"categories":1847},[],{"categories":1849},[157],{"categories":1851},[136],{"categories":1853},[113],{"categories":1855},[116],{"categories":1857},[136],{"categories":1859},[174],{"categories":1861},[],{"categories":1863},[],{"categories":1865},[160],{"categories":1867},[116,167],{"categories":1869},[136],{"categories":1871},[116],{"categories":1873},[64],{"categories":1875},[64],{"categories":1877},[116],{"categories":1879},[],{"categories":1881},[167],{"categories":1883},[116],{"categories":1885},[160],{"categories":1887},[64],{"categories":1889},[174],{"categories":1891},[273],{"categories":1893},[],{"categories":1895},[110],{"categories":1897},[64],{"categories":1899},[64],{"categories":1901},[167],{"categories":1903},[116],{"categories":1905},[116],{"categories":1907},[],{"categories":1909},[],{"categories":1911},[],{"categories":1913},[273],{"categories":1915},[136],{"categories":1917},[116],{"categories":1919},[116],{"categories":1921},[116],{"categories":1923},[],{"categories":1925},[160],{"categories":1927},[113],{"categories":1929},[],{"categories":1931},[64],{"categories":1933},[273],{"categories":1935},[],{"categories":1937},[157],{"categories":1939},[157],{"categories":1941},[],{"categories":1943},[167],{"categories":1945},[157],{"categories":1947},[116],{"categories":1949},[],{"categories":1951},[136],{"categories":1953},[116],{"categories":1955},[157],{"categories":1957},[64],{"categories":1959},[136],{"categories":1961},[],{"categories":1963},[64],{"categories":1965},[157],{"categories":1967},[116],{"categories":1969},[],{"categories":1971},[116],{"categories":1973},[116],{"categories":1975},[273],{"categories":1977},[136],{"categories":1979},[160],{"categories":1981},[160],{"categories":1983},[],{"categories":1985},[],{"categories":1987},[],{"categories":1989},[64],{"categories":1991},[167],{"categories":1993},[167],{"categories":1995},[],{"categories":1997},[],{"categories":1999},[116],{"categories":2001},[],{"categories":2003},[64],{"categories":2005},[116],{"categories":2007},[],{"categories":2009},[116],{"categories":2011},[113],{"categories":2013},[116],{"categories":2015},[174],{"categories":2017},[64],{"categories":2019},[116],{"categories":2021},[167],{"categories":2023},[],{"categories":2025},[136],{"categories":2027},[64],{"categories":2029},[],{"categories":2031},[136],{"categories":2033},[64],{"categories":2035},[64],{"categories":2037},[],{"categories":2039},[113],{"categories":2041},[64],{"categories":2043},[],{"categories":2045},[116],{"categories":2047},[110],{"categories":2049},[136],{"categories":2051},[273],{"categories":2053},[64],{"categories":2055},[64],{"categories":2057},[110],{"categories":2059},[],{"categories":2061},[116],{"categories":2063},[],{"categories":2065},[],{"categories":2067},[157],{"categories":2069},[116,113],{"categories":2071},[],{"categories":2073},[110],{"categories":2075},[160],{"categories":2077},[116],{"categories":2079},[167],{"categories":2081},[116],{"categories":2083},[64],{"categories":2085},[116],{"categories":2087},[116],{"categories":2089},[136],{"categories":2091},[64],{"categories":2093},[],{"categories":2095},[],{"categories":2097},[64],{"categories":2099},[116],{"categories":2101},[273],{"categories":2103},[],{"categories":2105},[116],{"categories":2107},[64],{"categories":2109},[],{"categories":2111},[116],{"categories":2113},[174],{"categories":2115},[160],{"categories":2117},[64],{"categories":2119},[116],{"categories":2121},[273],{"categories":2123},[],{"categories":2125},[116],{"categories":2127},[174],{"categories":2129},[157],{"categories":2131},[116],{"categories":2133},[],{"categories":2135},[174],{"categories":2137},[136],{"categories":2139},[116],{"categories":2141},[116],{"categories":2143},[110],{"categories":2145},[],{"categories":2147},[],{"categories":2149},[157],{"categories":2151},[116],{"categories":2153},[160],{"categories":2155},[174],{"categories":2157},[174],{"categories":2159},[136],{"categories":2161},[],{"categories":2163},[],{"categories":2165},[116],{"categories":2167},[],{"categories":2169},[116,167],{"categories":2171},[136],{"categories":2173},[64],{"categories":2175},[167],{"categories":2177},[116],{"categories":2179},[110],{"categories":2181},[],{"categories":2183},[],{"categories":2185},[110],{"categories":2187},[174],{"categories":2189},[116],{"categories":2191},[],{"categories":2193},[157,116],{"categories":2195},[273],{"categories":2197},[110],{"categories":2199},[],{"categories":2201},[113],{"categories":2203},[113],{"categories":2205},[116],{"categories":2207},[167],{"categories":2209},[64],{"categories":2211},[136],{"categories":2213},[174],{"categories":2215},[157],{"categories":2217},[116],{"categories":2219},[116],{"categories":2221},[116],{"categories":2223},[110],{"categories":2225},[116],{"categories":2227},[64],{"categories":2229},[136],{"categories":2231},[],{"categories":2233},[],{"categories":2235},[160],{"categories":2237},[167],{"categories":2239},[116],{"categories":2241},[157],{"categories":2243},[160],{"categories":2245},[116],{"categories":2247},[116],{"categories":2249},[64],{"categories":2251},[64],{"categories":2253},[116,113],{"categories":2255},[],{"categories":2257},[157],{"categories":2259},[],{"categories":2261},[116],{"categories":2263},[136],{"categories":2265},[110],{"categories":2267},[110],{"categories":2269},[64],{"categories":2271},[116],{"categories":2273},[113],{"categories":2275},[167],{"categories":2277},[174],{"categories":2279},[116],{"categories":2281},[],{"categories":2283},[136],{"categories":2285},[116],{"categories":2287},[116],{"categories":2289},[136],{"categories":2291},[167],{"categories":2293},[116],{"categories":2295},[64],{"categories":2297},[136],{"categories":2299},[116],{"categories":2301},[157],{"categories":2303},[116],{"categories":2305},[116],{"categories":2307},[273],{"categories":2309},[121],{"categories":2311},[64],{"categories":2313},[116],{"categories":2315},[136],{"categories":2317},[64],{"categories":2319},[174],{"categories":2321},[116],{"categories":2323},[],{"categories":2325},[116],{"categories":2327},[],{"categories":2329},[],{"categories":2331},[],{"categories":2333},[113],{"categories":2335},[116],{"categories":2337},[64],{"categories":2339},[136],{"categories":2341},[136],{"categories":2343},[136],{"categories":2345},[136],{"categories":2347},[],{"categories":2349},[110],{"categories":2351},[64],{"categories":2353},[136],{"categories":2355},[110],{"categories":2357},[64],{"categories":2359},[116],{"categories":2361},[116,64],{"categories":2363},[64],{"categories":2365},[273],{"categories":2367},[136],{"categories":2369},[136],{"categories":2371},[64],{"categories":2373},[116],{"categories":2375},[],{"categories":2377},[136],{"categories":2379},[174],{"categories":2381},[110],{"categories":2383},[116],{"categories":2385},[116],{"categories":2387},[],{"categories":2389},[167],{"categories":2391},[],{"categories":2393},[110],{"categories":2395},[64],{"categories":2397},[136],{"categories":2399},[116],{"categories":2401},[136],{"categories":2403},[110],{"categories":2405},[136],{"categories":2407},[136],{"categories":2409},[],{"categories":2411},[113],{"categories":2413},[64],{"categories":2415},[136],{"categories":2417},[136],{"categories":2419},[136],{"categories":2421},[136],{"categories":2423},[136],{"categories":2425},[136],{"categories":2427},[136],{"categories":2429},[136],{"categories":2431},[136],{"categories":2433},[136],{"categories":2435},[160],{"categories":2437},[110],{"categories":2439},[116],{"categories":2441},[116],{"categories":2443},[],{"categories":2445},[116,110],{"categories":2447},[],{"categories":2449},[64],{"categories":2451},[136],{"categories":2453},[64],{"categories":2455},[116],{"categories":2457},[116],{"categories":2459},[116],{"categories":2461},[116],{"categories":2463},[116],{"categories":2465},[64],{"categories":2467},[113],{"categories":2469},[157],{"categories":2471},[136],{"categories":2473},[116],{"categories":2475},[],{"categories":2477},[],{"categories":2479},[64],{"categories":2481},[157],{"categories":2483},[116],{"categories":2485},[],{"categories":2487},[],{"categories":2489},[174],{"categories":2491},[116],{"categories":2493},[],{"categories":2495},[],{"categories":2497},[110],{"categories":2499},[113],{"categories":2501},[116],{"categories":2503},[113],{"categories":2505},[157],{"categories":2507},[],{"categories":2509},[136],{"categories":2511},[],{"categories":2513},[157],{"categories":2515},[116],{"categories":2517},[174],{"categories":2519},[],{"categories":2521},[174],{"categories":2523},[],{"categories":2525},[],{"categories":2527},[64],{"categories":2529},[],{"categories":2531},[113],{"categories":2533},[110],{"categories":2535},[157],{"categories":2537},[167],{"categories":2539},[],{"categories":2541},[],{"categories":2543},[116],{"categories":2545},[110],{"categories":2547},[174],{"categories":2549},[],{"categories":2551},[64],{"categories":2553},[64],{"categories":2555},[136],{"categories":2557},[116],{"categories":2559},[64],{"categories":2561},[116],{"categories":2563},[64],{"categories":2565},[116],{"categories":2567},[121],{"categories":2569},[136],{"categories":2571},[],{"categories":2573},[174],{"categories":2575},[],{"categories":2577},[167],{"categories":2579},[64],{"categories":2581},[],{"categories":2583},[116],{"categories":2585},[64],{"categories":2587},[113],{"categories":2589},[110],{"categories":2591},[116],{"categories":2593},[157],{"categories":2595},[167],{"categories":2597},[167],{"categories":2599},[116],{"categories":2601},[160],{"categories":2603},[116],{"categories":2605},[64],{"categories":2607},[113],{"categories":2609},[64],{"categories":2611},[116],{"categories":2613},[116],{"categories":2615},[64],{"categories":2617},[136],{"categories":2619},[],{"categories":2621},[110],{"categories":2623},[116],{"categories":2625},[64],{"categories":2627},[116],{"categories":2629},[116],{"categories":2631},[],{"categories":2633},[157],{"categories":2635},[113],{"categories":2637},[136],{"categories":2639},[116],{"categories":2641},[116],{"categories":2643},[157],{"categories":2645},[174],{"categories":2647},[160],{"categories":2649},[116],{"categories":2651},[136],{"categories":2653},[116],{"categories":2655},[64],{"categories":2657},[273],{"categories":2659},[116],{"categories":2661},[64],{"categories":2663},[160],{"categories":2665},[],{"categories":2667},[64],{"categories":2669},[167],{"categories":2671},[157],{"categories":2673},[116],{"categories":2675},[110],{"categories":2677},[113],{"categories":2679},[167],{"categories":2681},[116],{"categories":2683},[],{"categories":2685},[64],{"categories":2687},[116],{"categories":2689},[],{"categories":2691},[136],{"categories":2693},[],{"categories":2695},[136],{"categories":2697},[116],{"categories":2699},[64],{"categories":2701},[64],{"categories":2703},[64],{"categories":2705},[],{"categories":2707},[],{"categories":2709},[116],{"categories":2711},[116],{"categories":2713},[],{"categories":2715},[157],{"categories":2717},[64],{"categories":2719},[174],{"categories":2721},[110],{"categories":2723},[],{"categories":2725},[],{"categories":2727},[136],{"categories":2729},[167],{"categories":2731},[116],{"categories":2733},[116],{"categories":2735},[116],{"categories":2737},[167],{"categories":2739},[136],{"categories":2741},[157],{"categories":2743},[116],{"categories":2745},[116],{"categories":2747},[116],{"categories":2749},[136],{"categories":2751},[116],{"categories":2753},[136],{"categories":2755},[136],{"categories":2757},[64],{"categories":2759},[64],{"categories":2761},[167],{"categories":2763},[136],{"categories":2765},[64],{"categories":2767},[116],{"categories":2769},[167],{"categories":2771},[157],{"categories":2773},[],{"categories":2775},[64],{"categories":2777},[],{"categories":2779},[],{"categories":2781},[],{"categories":2783},[113],{"categories":2785},[116],{"categories":2787},[64],{"categories":2789},[110],{"categories":2791},[64],{"categories":2793},[174],{"categories":2795},[],{"categories":2797},[64],{"categories":2799},[],{"categories":2801},[110],{"categories":2803},[64],{"categories":2805},[],{"categories":2807},[64],{"categories":2809},[116],{"categories":2811},[136],{"categories":2813},[116],{"categories":2815},[64],{"categories":2817},[136],{"categories":2819},[64],{"categories":2821},[167],{"categories":2823},[157],{"categories":2825},[110],{"categories":2827},[],{"categories":2829},[64],{"categories":2831},[157],{"categories":2833},[273],{"categories":2835},[136],{"categories":2837},[116],{"categories":2839},[157],{"categories":2841},[110],{"categories":2843},[],{"categories":2845},[64],{"categories":2847},[64],{"categories":2849},[116],{"categories":2851},[],{"categories":2853},[64],{"categories":2855},[121],{"categories":2857},[136],{"categories":2859},[64],{"categories":2861},[113],{"categories":2863},[],{"categories":2865},[116],{"categories":2867},[121],{"categories":2869},[116],{"categories":2871},[64],{"categories":2873},[136],{"categories":2875},[110],{"categories":2877},[273],{"categories":2879},[116],{"categories":2881},[116],{"categories":2883},[116],{"categories":2885},[136],{"categories":2887},[113],{"categories":2889},[116],{"categories":2891},[157],{"categories":2893},[136],{"categories":2895},[273],{"categories":2897},[116],{"categories":2899},[],{"categories":2901},[],{"categories":2903},[116],{"categories":2905},[273],{"categories":2907},[160],{"categories":2909},[64],{"categories":2911},[64],{"categories":2913},[136],{"categories":2915},[116],{"categories":2917},[110],{"categories":2919},[157],{"categories":2921},[64],{"categories":2923},[116],{"categories":2925},[174],{"categories":2927},[116],{"categories":2929},[64],{"categories":2931},[],{"categories":2933},[116],{"categories":2935},[116],{"categories":2937},[136],{"categories":2939},[110],{"categories":2941},[],{"categories":2943},[116],{"categories":2945},[116],{"categories":2947},[167],{"categories":2949},[157],{"categories":2951},[116,64],{"categories":2953},[174,113],{"categories":2955},[116],{"categories":2957},[],{"categories":2959},[64],{"categories":2961},[],{"categories":2963},[167],{"categories":2965},[],{"categories":2967},[116],{"categories":2969},[136],{"categories":2971},[],{"categories":2973},[64],{"categories":2975},[],{"categories":2977},[157],{"categories":2979},[64],{"categories":2981},[110],{"categories":2983},[64],{"categories":2985},[116],{"categories":2987},[],{"categories":2989},[273],{"categories":2991},[174],{"categories":2993},[113],{"categories":2995},[113],{"categories":2997},[110],{"categories":2999},[110],{"categories":3001},[116],{"categories":3003},[64],{"categories":3005},[116],{"categories":3007},[116],{"categories":3009},[110],{"categories":3011},[116],{"categories":3013},[174],{"categories":3015},[136],{"categories":3017},[116],{"categories":3019},[64],{"categories":3021},[116],{"categories":3023},[],{"categories":3025},[167],{"categories":3027},[],{"categories":3029},[64],{"categories":3031},[110],{"categories":3033},[],{"categories":3035},[273],{"categories":3037},[116],{"categories":3039},[],{"categories":3041},[136],{"categories":3043},[64],{"categories":3045},[167],{"categories":3047},[116],{"categories":3049},[64],{"categories":3051},[167],{"categories":3053},[64],{"categories":3055},[136],{"categories":3057},[110],{"categories":3059},[136],{"categories":3061},[167],{"categories":3063},[116],{"categories":3065},[157],{"categories":3067},[116],{"categories":3069},[116],{"categories":3071},[116],{"categories":3073},[116],{"categories":3075},[64],{"categories":3077},[116],{"categories":3079},[64],{"categories":3081},[116],{"categories":3083},[110],{"categories":3085},[116],{"categories":3087},[64],{"categories":3089},[157],{"categories":3091},[110],{"categories":3093},[64],{"categories":3095},[157],{"categories":3097},[],{"categories":3099},[116],{"categories":3101},[116],{"categories":3103},[167],{"categories":3105},[],{"categories":3107},[64],{"categories":3109},[174],{"categories":3111},[116],{"categories":3113},[136],{"categories":3115},[174],{"categories":3117},[64],{"categories":3119},[113],{"categories":3121},[113],{"categories":3123},[116],{"categories":3125},[110],{"categories":3127},[],{"categories":3129},[116],{"categories":3131},[],{"categories":3133},[110],{"categories":3135},[116],{"categories":3137},[64],{"categories":3139},[64],{"categories":3141},[],{"categories":3143},[167],{"categories":3145},[167],{"categories":3147},[174],{"categories":3149},[157],{"categories":3151},[],{"categories":3153},[116],{"categories":3155},[110],{"categories":3157},[116],{"categories":3159},[167],{"categories":3161},[110],{"categories":3163},[136],{"categories":3165},[136],{"categories":3167},[],{"categories":3169},[136],{"categories":3171},[64],{"categories":3173},[157],{"categories":3175},[160],{"categories":3177},[116],{"categories":3179},[],{"categories":3181},[136],{"categories":3183},[167],{"categories":3185},[113],{"categories":3187},[116],{"categories":3189},[110],{"categories":3191},[273],{"categories":3193},[110],{"categories":3195},[],{"categories":3197},[],{"categories":3199},[136],{"categories":3201},[],{"categories":3203},[64],{"categories":3205},[64],{"categories":3207},[64],{"categories":3209},[],{"categories":3211},[116],{"categories":3213},[],{"categories":3215},[136],{"categories":3217},[110],{"categories":3219},[157],{"categories":3221},[116],{"categories":3223},[136],{"categories":3225},[136],{"categories":3227},[],{"categories":3229},[136],{"categories":3231},[110],{"categories":3233},[116],{"categories":3235},[],{"categories":3237},[64],{"categories":3239},[64],{"categories":3241},[110],{"categories":3243},[],{"categories":3245},[],{"categories":3247},[],{"categories":3249},[157],{"categories":3251},[64],{"categories":3253},[116],{"categories":3255},[],{"categories":3257},[],{"categories":3259},[],{"categories":3261},[157],{"categories":3263},[],{"categories":3265},[110],{"categories":3267},[],{"categories":3269},[],{"categories":3271},[157],{"categories":3273},[116],{"categories":3275},[136],{"categories":3277},[],{"categories":3279},[174],{"categories":3281},[136],{"categories":3283},[174],{"categories":3285},[116],{"categories":3287},[],{"categories":3289},[],{"categories":3291},[64],{"categories":3293},[],{"categories":3295},[],{"categories":3297},[64],{"categories":3299},[116],{"categories":3301},[],{"categories":3303},[64],{"categories":3305},[136],{"categories":3307},[174],{"categories":3309},[160],{"categories":3311},[64],{"categories":3313},[64],{"categories":3315},[],{"categories":3317},[],{"categories":3319},[],{"categories":3321},[136],{"categories":3323},[],{"categories":3325},[],{"categories":3327},[157],{"categories":3329},[110],{"categories":3331},[],{"categories":3333},[113],{"categories":3335},[174],{"categories":3337},[116],{"categories":3339},[167],{"categories":3341},[110],{"categories":3343},[160],{"categories":3345},[113],{"categories":3347},[167],{"categories":3349},[],{"categories":3351},[],{"categories":3353},[64],{"categories":3355},[110],{"categories":3357},[157],{"categories":3359},[110],{"categories":3361},[64],{"categories":3363},[273],{"categories":3365},[116],{"categories":3367},[110],{"categories":3369},[64],{"categories":3371},[],{"categories":3373},[116],{"categories":3375},[136],{"categories":3377},[167],{"categories":3379},[],{"categories":3381},[157],{"categories":3383},[136],{"categories":3385},[110],{"categories":3387},[64],{"categories":3389},[116],{"categories":3391},[113],{"categories":3393},[64,273],{"categories":3395},[64],{"categories":3397},[167],{"categories":3399},[116],{"categories":3401},[160],{"categories":3403},[174],{"categories":3405},[64],{"categories":3407},[],{"categories":3409},[64],{"categories":3411},[116],{"categories":3413},[113],{"categories":3415},[],{"categories":3417},[],{"categories":3419},[116],{"categories":3421},[160],{"categories":3423},[116],{"categories":3425},[],{"categories":3427},[136],{"categories":3429},[],{"categories":3431},[136],{"categories":3433},[167],{"categories":3435},[64],{"categories":3437},[116],{"categories":3439},[174],{"categories":3441},[167],{"categories":3443},[],{"categories":3445},[136],{"categories":3447},[116],{"categories":3449},[],{"categories":3451},[116],{"categories":3453},[64],{"categories":3455},[116],{"categories":3457},[64],{"categories":3459},[116],{"categories":3461},[116],{"categories":3463},[116],{"categories":3465},[116],{"categories":3467},[113],{"categories":3469},[],{"categories":3471},[121],{"categories":3473},[136],{"categories":3475},[116],{"categories":3477},[],{"categories":3479},[167],{"categories":3481},[116],{"categories":3483},[116],{"categories":3485},[116],{"categories":3487},[64],{"categories":3489},[136],{"categories":3491},[116],{"categories":3493},[116],{"categories":3495},[113],{"categories":3497},[64],{"categories":3499},[157],{"categories":3501},[],{"categories":3503},[160],{"categories":3505},[116],{"categories":3507},[],{"categories":3509},[136],{"categories":3511},[174],{"categories":3513},[],{"categories":3515},[],{"categories":3517},[136],{"categories":3519},[136],{"categories":3521},[174],{"categories":3523},[110],{"categories":3525},[64],{"categories":3527},[64],{"categories":3529},[116],{"categories":3531},[113],{"categories":3533},[],{"categories":3535},[],{"categories":3537},[136],{"categories":3539},[160],{"categories":3541},[167],{"categories":3543},[64],{"categories":3545},[157],{"categories":3547},[160],{"categories":3549},[160],{"categories":3551},[],{"categories":3553},[136],{"categories":3555},[116],{"categories":3557},[116],{"categories":3559},[167],{"categories":3561},[],{"categories":3563},[136],{"categories":3565},[136],{"categories":3567},[136],{"categories":3569},[],{"categories":3571},[64],{"categories":3573},[116],{"categories":3575},[],{"categories":3577},[110],{"categories":3579},[113],{"categories":3581},[],{"categories":3583},[116],{"categories":3585},[116],{"categories":3587},[],{"categories":3589},[167],{"categories":3591},[],{"categories":3593},[],{"categories":3595},[],{"categories":3597},[],{"categories":3599},[116],{"categories":3601},[136],{"categories":3603},[],{"categories":3605},[],{"categories":3607},[116],{"categories":3609},[116],{"categories":3611},[116],{"categories":3613},[160],{"categories":3615},[116],{"categories":3617},[160],{"categories":3619},[],{"categories":3621},[160],{"categories":3623},[160],{"categories":3625},[273],{"categories":3627},[64],{"categories":3629},[167],{"categories":3631},[],{"categories":3633},[],{"categories":3635},[160],{"categories":3637},[167],{"categories":3639},[167],{"categories":3641},[167],{"categories":3643},[],{"categories":3645},[110],{"categories":3647},[167],{"categories":3649},[167],{"categories":3651},[110],{"categories":3653},[167],{"categories":3655},[113],{"categories":3657},[167],{"categories":3659},[167],{"categories":3661},[167],{"categories":3663},[160],{"categories":3665},[136],{"categories":3667},[136],{"categories":3669},[116],{"categories":3671},[167],{"categories":3673},[160],{"categories":3675},[273],{"categories":3677},[160],{"categories":3679},[160],{"categories":3681},[160],{"categories":3683},[],{"categories":3685},[113],{"categories":3687},[],{"categories":3689},[273],{"categories":3691},[167],{"categories":3693},[167],{"categories":3695},[167],{"categories":3697},[64],{"categories":3699},[136,113],{"categories":3701},[160],{"categories":3703},[],{"categories":3705},[],{"categories":3707},[160],{"categories":3709},[],{"categories":3711},[160],{"categories":3713},[136],{"categories":3715},[64],{"categories":3717},[],{"categories":3719},[167],{"categories":3721},[116],{"categories":3723},[157],{"categories":3725},[],{"categories":3727},[116],{"categories":3729},[],{"categories":3731},[136],{"categories":3733},[110],{"categories":3735},[160],{"categories":3737},[],{"categories":3739},[167],{"categories":3741},[136],[3743,3825,4050,4136],{"id":3744,"title":3745,"ai":3746,"body":3751,"categories":3779,"created_at":65,"date_modified":65,"description":57,"extension":66,"faq":65,"featured":67,"kicker_label":65,"meta":3780,"navigation":86,"path":3810,"published_at":3811,"question":65,"scraped_at":3812,"seo":3813,"sitemap":3814,"source_id":3815,"source_name":93,"source_type":94,"source_url":3816,"stem":3817,"tags":3818,"thumbnail_url":3820,"tldr":3821,"tweet":3822,"unknown_tags":3823,"__hash__":3824},"summaries\u002Fsummaries\u002F13a7be7216932e4a-agentic-data-cloud-powers-ai-swarms-from-insights--summary.md","Agentic Data Cloud Powers AI Swarms from Insights to Action",{"provider":7,"model":8,"input_tokens":3747,"output_tokens":3748,"processing_time_ms":3749,"cost_usd":3750},8084,2064,37170,0.00262615,{"type":14,"value":3752,"toc":3774},[3753,3757,3760,3764,3767,3771],[17,3754,3756],{"id":3755},"enrich-data-with-inferred-context-for-accurate-agent-reasoning","Enrich Data with Inferred Context for Accurate Agent Reasoning",[22,3758,3759],{},"Traditional data strategies focused on clean, structured data with lineage and quality, achieving only 50% agent accuracy; the other 50% requires business context like hidden PDF codes or supply chain meanings previously siloed in human minds. Google's Knowledge Catalog uses GenAI to infer schemas, descriptions, and relationships across thousands of structured\u002Funstructured files (e.g., PDFs), avoiding context window limits and high costs of feeding raw docs to models. This enables agents to access precise, low-cost context, boosting trust and efficiency over human-coded glossaries. Combined with semantic hybrid search from Google Search (including reranking), it serves optimal real-time context, preventing agents from getting lost in excess data.",[17,3761,3763],{"id":3762},"intent-driven-agent-swarms-via-tools-and-integrations","Intent-Driven Agent Swarms via Tools and Integrations",[22,3765,3766],{},"Move beyond persona-based agents to swarms handling end-to-end tasks (data wrangling, modeling, visualization, deployment) with intent-driven engineering, freeing practitioners for outcomes over tasks. Data Agent Kit provides pre-built plugins\u002Fskills for natively orchestrating BigQuery, Spark, AlloyDB pipelines, and multi-tool actions into ledgers\u002Fmarketing systems. Gemini Enterprise acts as the front door: business users chat with conversational agents backed by BigQuery\u002FLooker assets or Deep Research agents blending enterprise data (via Knowledge Catalog) with web\u002Fdocs for holistic insights (e.g., shipping optimization from weather\u002Ftraffic + internal data) in seconds vs. weeks. This grounds agents in machine-readable, real-time data for true ROI.",[17,3768,3770],{"id":3769},"multi-cloud-scale-with-optimized-economics","Multi-Cloud Scale with Optimized Economics",[22,3772,3773],{},"Apache Iceberg open standard and Cross-Cloud Lakehouse unify data across AWS S3, Azure, Databricks Unity Catalog, Snowflake without migration, using Cross-Cloud Interconnect for subsecond latencies and petabyte-scale transfers, solving proprietary format silos. For agent-scale (10-20 API calls per human click, swarms spiking traffic), Google optimizes the full stack: BigQuery 35% faster queries\u002F40% cost drop, Spark Lightning Engine 5x faster\u002F2x price-performance vs. alternatives, 230x token reduction for AI inferencing. TPU v8 separates training\u002Finference to avoid silicon bottlenecks. Vertical integration ensures seamless, cost-effective low-latency swarms, turning 45-minute human processes into 1-minute actions.",{"title":57,"searchDepth":58,"depth":58,"links":3775},[3776,3777,3778],{"id":3755,"depth":58,"text":3756},{"id":3762,"depth":58,"text":3763},{"id":3769,"depth":58,"text":3770},[64],{"content_references":3781,"triage":3807},[3782,3785,3787,3789,3791,3793,3795,3798,3800,3802,3804],{"type":71,"title":3783,"author":3784,"context":74},"Agentic Data Cloud","Google Cloud",{"type":71,"title":3786,"author":3784,"context":74},"Google Cloud Data Agent Kit",{"type":71,"title":3788,"author":3784,"context":74},"Knowledge Catalog",{"type":71,"title":3790,"author":3784,"context":74},"BigQuery",{"type":71,"title":3792,"author":3784,"context":74},"Spark",{"type":71,"title":3794,"context":74},"Apache Iceberg",{"type":71,"title":3796,"author":3797,"context":74},"Gemini Enterprise","Google",{"type":71,"title":3799,"author":3784,"context":74},"Deep Research Agent",{"type":71,"title":3801,"author":3784,"context":74},"Cross Cloud Lakehouse",{"type":71,"title":3803,"author":3784,"context":74},"TPU v8",{"type":3805,"title":3806,"author":3784,"context":74},"event","Google Cloud Next '26",{"relevance":82,"novelty":83,"quality":82,"actionability":83,"composite":3808,"reasoning":3809},3.6,"Category: AI Automation. The article discusses the use of context-enriched data and agent swarms for improved AI functionality, addressing the audience's need for practical applications of AI tools. It provides insights into tools like the Data Agent Kit and Knowledge Catalog, which can be directly relevant for product builders looking to implement AI solutions.","\u002Fsummaries\u002F13a7be7216932e4a-agentic-data-cloud-powers-ai-swarms-from-insights-summary","2026-05-14 04:00:36","2026-05-14 07:00:48",{"title":3745,"description":57},{"loc":3810},"13a7be7216932e4a","https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=WQIODOuXWr8","summaries\u002F13a7be7216932e4a-agentic-data-cloud-powers-ai-swarms-from-insights--summary",[98,99,3819,100],"automation","https:\u002F\u002Fi.ytimg.com\u002Fvi\u002FWQIODOuXWr8\u002Fhqdefault.jpg","Shift data platforms from systems of intelligence (1-20% insights actioned) to action via context-enriched data in Knowledge Catalog, Data Agent Kit tools for BigQuery\u002FSpark, and infra optimizations like 230x token cuts for efficient agent swarms.","Google Cloud exec Yasmeen Ahmad pitches their \"Agentic Data Cloud\" in a conference interview, explaining Knowledge Catalog for data context, Data Agent Kit for agent tools on BigQuery\u002FSpark, and Iceberg for multi-cloud access. Mostly high-level vision talk with customer anecdotes.",[],"akSdWwtwyi_3coecpdJV_HzQg8hZ9t0z00xizfgzO0Q",{"id":3826,"title":3827,"ai":3828,"body":3833,"categories":4017,"created_at":65,"date_modified":65,"description":57,"extension":66,"faq":65,"featured":67,"kicker_label":65,"meta":4018,"navigation":86,"path":4034,"published_at":4035,"question":65,"scraped_at":4036,"seo":4037,"sitemap":4038,"source_id":4039,"source_name":4040,"source_type":94,"source_url":4041,"stem":4042,"tags":4043,"thumbnail_url":4045,"tldr":4046,"tweet":4047,"unknown_tags":4048,"__hash__":4049},"summaries\u002Fsummaries\u002F9732ce9fa72d407a-mind-the-gap-observability-for-drifting-ai-agents-summary.md","Mind the Gap: Observability for Drifting AI Agents",{"provider":7,"model":8,"input_tokens":3829,"output_tokens":3830,"processing_time_ms":3831,"cost_usd":3832},8792,2878,35833,0.00317475,{"type":14,"value":3834,"toc":4009},[3835,3839,3842,3845,3848,3852,3855,3858,3874,3877,3880,3883,3887,3890,3911,3914,3917,3920,3924,3927,3948,3951,3954,3957,3960,3964,3967,3970,3973,3976,3980],[17,3836,3838],{"id":3837},"agent-drift-creates-invisible-gapsobservability-closes-them","Agent Drift Creates Invisible Gaps—Observability Closes Them",[22,3840,3841],{},"Agents degrade silently: models update, prompts tweak, edge cases pile up, widening the divide between intended behavior (the \"platform\") and reality (the \"train\"). Amy Boyd and Nitya Narasimhan analogize this to London's \"Mind the Gap\" signs, where platforms stay fixed but trains evolve, risking falls without constant checks. They advocate observability across the lifecycle—build, debug\u002Fproduction, multi-agent fleets—to measure quality, safety, and agentic performance like intent resolution and task adherence.",[22,3843,3844],{},"Key decision: Start evaluations early, not as an afterthought. Without baselines or datasets (common for new agents), manual setup wastes time. Tradeoff: Non-determinism means scores are probabilistic (e.g., 80-90% tool call accuracy), not binary, requiring continuous monitoring over snapshots.",[22,3846,3847],{},"\"Agents are non-deterministic. That's not just a problem for demos. That's also a problem for real life when you actually get to production.\" —Amy Boyd, emphasizing why reliability demands tracing, evals, monitoring, and optimization, not just one-off tests.",[17,3849,3851],{"id":3850},"tracing-with-opentelemetry-unifies-multi-tool-agent-workflows","Tracing with OpenTelemetry Unifies Multi-Tool Agent Workflows",[22,3853,3854],{},"Build observability from day zero using OpenTelemetry (OTel) standards for tracing. This captures full agent traces: tool calls, messages, decisions across workflows—even multi-agent systems where debugging explodes in complexity.",[22,3856,3857],{},"In Microsoft Foundry, instrument agents built anywhere (e.g., LangChain, Semantic Kernel) and centralize traces in the control plane. No vendor lock-in: OTel exports to Foundry or your tools. For a weather agent example:",[3859,3860,3861,3865,3868,3871],"ul",{},[3862,3863,3864],"li",{},"User query: \"What's the weather in London?\"",[3862,3866,3867],{},"Eval 1: Intent resolution (did it detect local weather request?).",[3862,3869,3870],{},"Eval 2: Tool call correctness (right API? Parameters match?).",[3862,3872,3873],{},"Eval 3: Task adherence\u002Fresponse quality.",[22,3875,3876],{},"This pinpoints failures: 95% intent match but 70% wrong tool? Tweak prompt there. Results: Developers debug faster, IT admins integrate with Azure Monitor for infra signals.",[22,3878,3879],{},"Tradeoffs: Traces balloon in multi-agent setups (exponential data), so Foundry aggregates fleet-wide views. Fork their GitHub repo (with dev containers for instant Codespaces setup) to replicate: pre-installed tools, notebooks, branches for evolving workshops.",[22,3881,3882],{},"\"If you think about the change in trains and technology, but the platform doesn't change. So there's this sign on there because at each station it's different.\" —Amy Boyd, illustrating how fixed requirements clash with evolving agents, solved by per-step tracing.",[17,3884,3886],{"id":3885},"built-in-and-custom-evaluators-for-quality-safety-agentics","Built-in and Custom Evaluators for Quality, Safety, Agentics",[22,3888,3889],{},"Foundry embeds 20+ evaluators, learned from platform-scale customer data:",[3859,3891,3892,3899,3905],{},[3862,3893,3894,3898],{},[3895,3896,3897],"strong",{},"Quality",": Coherence, relevance, groundedness.",[3862,3900,3901,3904],{},[3895,3902,3903],{},"Safety\u002FRisk",": Toxicity, bias, PII leakage.",[3862,3906,3907,3910],{},[3895,3908,3909],{},"Agentic",": Intent resolution, tool selection, task completion—holistic over single LLM calls.",[22,3912,3913],{},"No eval data? Custom evaluators chain LLMs (e.g., GPT-4o judges outputs). Run continuous (code changes), scheduled, or ad-hoc batches. Metrics: Percentages reveal drift (e.g., task adherence drops from 92% to 75% post-prompt tweak).",[22,3915,3916],{},"For new projects like a Contoso travel agent (hotels, cars, flights): Too many models (2M+ on Hugging Face, 11K+ in Azure)? No data? Foundry prototypes fast: Define instructions\u002Ftools\u002Fmodel, trace first run, eval workflow.",[22,3918,3919],{},"\"You have no existing data. This app never existed before. Where do I get the data to even do evals?\" —Nitya Narasimhan, highlighting the zero-to-prototype challenge solved by auto-generated datasets in their observe skill.",[17,3921,3923],{"id":3922},"observe-skill-one-prompt-auto-evals-optimization-rollback","Observe Skill: One-Prompt Auto-Evals, Optimization, Rollback",[22,3925,3926],{},"Demo steals the show: \"Observe skill\"—a meta-agent for agents with zero setup.",[3928,3929,3930,3933,3936,3939,3942,3945],"ol",{},[3862,3931,3932],{},"Point at your agent (e.g., coding agent).",[3862,3934,3935],{},"Generates synthetic eval dataset (queries, gold responses).",[3862,3937,3938],{},"Batch evals across metrics.",[3862,3940,3941],{},"A\u002FB tests prompt versions.",[3862,3943,3944],{},"Optimizes (e.g., rephrases for +15% task adherence).",[3862,3946,3947],{},"Rolls back to best; shows reasoning\u002Ftrace.",[22,3949,3950],{},"In travel agent: Surfaces unknown failures like hallucinated bookings or wrong tool chains. Timeline: Minutes, not days. Costs: ~$10 for workshop runs (Azure free tier viable). Repo includes Codespaces dev container: VS Code in browser, deps pre-installed, extensions for notebooks\u002Fskills.",[22,3952,3953],{},"Tradeoffs: Relies on judge-model quality (use strong ones like o1); synthetic data misses real edges (pair with red teaming). Evolution: v1 single-agent → multi-agent traces.",[22,3955,3956],{},"This accelerates optimize loop: Data → Insight → Iteration. Nitya: Treat repo as 4-hour workshop; fork all branches for updates.",[22,3958,3959],{},"\"The skill shows its reasoning at each step, which is where the real value is: it surfaces the failures you didn't know to look for.\" —From session description, underscoring transparent auto-debugging over black-box scores.",[17,3961,3963],{"id":3962},"red-teaming-and-fleet-scale-monitoring-for-production","Red Teaming and Fleet-Scale Monitoring for Production",[22,3965,3966],{},"Safety beyond normal users: Red teaming pits adversarial AI (second agent) against yours with jailbreak prompts, finding vulns pre-launch. Integrates open-source like Pirat repo; one-click in Foundry.",[22,3968,3969],{},"Scale to fleets: Centralized dashboard across many multi-agent systems. Pull cloud monitors (Azure), schedule evals, fleet views. Hosting-agnostic: Build in LlamaIndex, host\u002Fobserve in Foundry.",[22,3971,3972],{},"Results: Brand protection (quality), user safety (guardrails), quick fixes (e.g., 40% faster debug via traces). Future: Security deep-dive (missed here, but repo\u002FDiscord cover).",[22,3974,3975],{},"\"Red teaming is not something you do alone.\" —Amy Boyd, on collaborating via open-source and platform tools to stress-test agents proactively.",[17,3977,3979],{"id":3978},"key-takeaways","Key Takeaways",[3859,3981,3982,3985,3988,3991,3994,3997,4000,4003,4006],{},[3862,3983,3984],{},"Fork the Microsoft Foundry repo early; use Codespaces dev containers to skip env setup and run 4-hour workshops in minutes.",[3862,3986,3987],{},"Instrument all agents with OpenTelemetry for portable tracing—centralize in Foundry control plane regardless of build tools.",[3862,3989,3990],{},"Evaluate workflows end-to-end: Intent → Tool call → Task adherence; use built-ins first, customize for gaps.",[3862,3992,3993],{},"Deploy 'observe skill' for zero-data starts: Auto-generate evals, optimize prompts, rollback—watch reasoning traces.",[3862,3995,3996],{},"Red team adversarially; monitor fleets continuously to catch drift before users notice.",[3862,3998,3999],{},"Start small: Prototype travel\u002Fcoding agents in repo to learn model\u002Ftools\u002Finstructions triad.",[3862,4001,4002],{},"Budget ~$10\u002Frun; join Discord for credits\u002Ftips—non-deterministic agents demand probabilistic metrics (80-95% norms).",[3862,4004,4005],{},"Tradeoff honesty: Synthetics great for baselines, but pair with real\u002Fuser data for production.",[3862,4007,4008],{},"Evolve: Single → Multi-agent → Fleet; OTel ensures future-proofing.",{"title":57,"searchDepth":58,"depth":58,"links":4010},[4011,4012,4013,4014,4015,4016],{"id":3837,"depth":58,"text":3838},{"id":3850,"depth":58,"text":3851},{"id":3885,"depth":58,"text":3886},{"id":3922,"depth":58,"text":3923},{"id":3962,"depth":58,"text":3963},{"id":3978,"depth":58,"text":3979},[],{"content_references":4019,"triage":4030},[4020,4022,4024,4028],{"type":71,"title":4021,"context":74},"OpenTelemetry",{"type":71,"title":4023,"context":74},"Azure Monitor",{"type":4025,"title":4026,"context":4027},"other","Microsoft Foundry GitHub Repo","recommended",{"type":4025,"title":4029,"context":74},"Pirat Open-Source Repository",{"relevance":4031,"novelty":82,"quality":82,"actionability":82,"composite":4032,"reasoning":4033},5,4.35,"Category: AI & LLMs. The article discusses observability for AI agents, addressing a critical pain point of agent drift and providing practical insights on using OpenTelemetry for monitoring and optimizing AI workflows. It offers concrete examples of evaluations and metrics that developers can implement, making it highly actionable.","\u002Fsummaries\u002F9732ce9fa72d407a-mind-the-gap-observability-for-drifting-ai-agents-summary","2026-05-14 16:00:06","2026-05-14 19:00:15",{"title":3827,"description":57},{"loc":4034},"9732ce9fa72d407a","AI Engineer","https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=iOXM3zE-2dk","summaries\u002F9732ce9fa72d407a-mind-the-gap-observability-for-drifting-ai-agents-summary",[98,99,4044,101],"dev-productivity","https:\u002F\u002Fi.ytimg.com\u002Fvi\u002FiOXM3zE-2dk\u002Fhqdefault.jpg","Microsoft Foundry's stack uses OpenTelemetry tracing, built-in evaluators, red teaming, and an 'observe skill' to detect agent drift, evaluate workflows, and auto-optimize prompts—bridging expected vs. actual behavior from build to production.","Microsoft devs demo Foundry's agent observability tools: OpenTelemetry tracing, built-in evaluators for quality\u002Fsafety\u002Fintent\u002Ftask metrics, AI red teaming with adversarial prompts. Standout is the \"observe skill\" demo, which auto-generates eval datasets, runs batch evals, prompt-optimizes, version-compares, and auto-rolls back—all from one prompt, showing its step-by-step reasoning.",[4044,101],"G4ggGZuI_N4D07UGBWZ7GMZvFtKoVS6Xrz4QJwiMS2E",{"id":4051,"title":4052,"ai":4053,"body":4058,"categories":4086,"created_at":65,"date_modified":65,"description":57,"extension":66,"faq":65,"featured":67,"kicker_label":65,"meta":4087,"navigation":86,"path":4120,"published_at":4121,"question":65,"scraped_at":4122,"seo":4123,"sitemap":4124,"source_id":4125,"source_name":4126,"source_type":94,"source_url":4127,"stem":4128,"tags":4129,"thumbnail_url":4131,"tldr":4132,"tweet":4133,"unknown_tags":4134,"__hash__":4135},"summaries\u002Fsummaries\u002F26831750495fa9ed-openai-s-real-time-voice-ai-powers-agents-backed-b-summary.md","OpenAI's Real-Time Voice AI Powers Agents, Backed by MRC Networking",{"provider":7,"model":8,"input_tokens":4054,"output_tokens":4055,"processing_time_ms":4056,"cost_usd":4057},6837,2244,28673,0.00246675,{"type":14,"value":4059,"toc":4081},[4060,4064,4067,4071,4074,4078],[17,4061,4063],{"id":4062},"real-time-voice-models-enable-production-agents","Real-Time Voice Models Enable Production Agents",[22,4065,4066],{},"OpenAI's GPT-Realtime-2 combines GPT-4o-class reasoning with low-latency voice for live agents handling complex tasks like flight rebooking under $400 by querying accounts, comparing options, issuing refunds, and explaining in parallel via multiple tools. It uses short filler phrases (\"Let me check\") to mimic human pauses, preventing awkward silences, and supports adjustable reasoning levels (minimal to X-high) for speed vs. depth—default low prioritizes \u003C500ms responses. Context window expands to 128k tokens from 32k, sustaining long support calls or tutoring. Benchmarks: 96.6% Big Bench Audio accuracy (vs. 81.4% prior), 48.5% Audio Multi-Challenge pass rate (vs. 34.7%). Handles interruptions, accents, medical terms, and tone shifts (calm, empathetic). GPT-Realtime-Translate supports 70+ input\u002F13 output languages with context-aware live translation for support or events (e.g., Deutsche Telekom testing). GPT-Realtime-Whisper streams transcription for captions, notes, and action items. Pricing: GPT-Realtime-2 at $32\u002FM input tokens ($0.40\u002FM cached), $64\u002FM output; Translate $0.034\u002Fmin; Whisper $0.017\u002Fmin. Patterns: voice-to-action (tools), systems-to-voice (app guidance), voice-to-voice (translation). EU residency and anti-spam guardrails included.",[17,4068,4070],{"id":4069},"mrc-networking-scales-frontier-training","MRC Networking Scales Frontier Training",[22,4072,4073],{},"MRC (Multi-Path Reliable Connection) optimizes GPU clusters by spreading data across hundreds of paths using RoCE\u002FRDMA and SRV6 routing, reducing bottlenecks vs. single-path systems. Failure recovery happens in microseconds at NIC level—e.g., reroutes around bad links without crashing jobs, restoring capacity post-failure in ~1 minute. Enables 131k GPUs with 2 switch tiers (vs. 3-4), using 2\u002F3 optics and 3\u002F5 switches, cutting latency. Handles 400\u002F800Gbit RDMA cards (Nvidia\u002FAMD\u002FBroadcom) and switches (Nvidia Spectrum\u002FBroadcom Tomahawk). Live on OpenAI's GB200 clusters (Oracle Abilene, Microsoft Fairwater), survived switch reboots mid-training for ChatGPT\u002Fo1 models. Trade-off: Shifts AI race from GPUs to networks, as idle time on $expensive hardware burns cash for 900M weekly ChatGPT users.",[17,4075,4077],{"id":4076},"ai-jobs-debate-washing-vs-real-displacement","AI Jobs Debate: Washing vs. Real Displacement",[22,4079,4080],{},"Sam Altman notes \"AI-washing\"—firms blame unrelated layoffs (margins, consumers, geopolitics) on AI to justify spending. Yet displacement grows: Anthropic's Amodei predicts 50% entry-level office jobs lost; Snap cut 16% citing AI; WEF says 40% employers plan reductions. Data mixed—NBER survey: 90% execs report no employment impact post-ChatGPT; Yale Budget Lab: no occupation shifts\u002Funemployment spikes through Mar 2026. Contrasts: 2.7% YoY productivity jump (Stanford); 13% employment drop for early-career AI-exposed roles. Analogy: Like computers, AI effects lag macro data. Outcome: Entry digital tasks shrink first; experienced roles stable\u002Fgrow.",{"title":57,"searchDepth":58,"depth":58,"links":4082},[4083,4084,4085],{"id":4062,"depth":58,"text":4063},{"id":4069,"depth":58,"text":4070},{"id":4076,"depth":58,"text":4077},[136],{"content_references":4088,"triage":4117},[4089,4093,4096,4099,4103,4107,4110,4114],{"type":4025,"title":4090,"url":4091,"context":4092},"Advancing voice intelligence with new models in the API","https:\u002F\u002Fopenai.com\u002Findex\u002Fadvancing-voice-intelligence-with-new-models-in-the-api\u002F","cited",{"type":4025,"title":4094,"url":4095,"context":4092},"Realtime agents guide","https:\u002F\u002Fdevelopers.openai.com\u002Fcookbook\u002Fexamples\u002Fvoice_solutions\u002Frealtime_agents_guide",{"type":4025,"title":4097,"url":4098,"context":4092},"MRC supercomputer networking","https:\u002F\u002Fopenai.com\u002Findex\u002Fmrc-supercomputer-networking\u002F",{"type":4100,"title":4101,"url":4102,"context":4092},"report","Sam Altman says some companies are AI-washing","https:\u002F\u002Fwww.techradar.com\u002Fpro\u002Fsam-altman-says-some-companies-are-ai-washing-by-blaming-unrelated-layoffs-on-the-technology-but-admits-things-may-get-worse-soon",{"type":4100,"title":4104,"publisher":4105,"url":4106,"context":4092},"Tracking impact of AI on labor market","Yale Budget Lab","https:\u002F\u002Fbudgetlab.yale.edu\u002Fresearch\u002Ftracking-impact-ai-labor-market",{"type":71,"title":4108,"url":4109,"context":4027},"CodeRabbit","https:\u002F\u002Fcoderabbit.link\u002Fai-revolution",{"type":4111,"title":4112,"author":4113,"context":4092},"paper","National Bureau of Economic Research study","National Bureau of Economic Research",{"type":4100,"title":4115,"publisher":4116,"context":4092},"World Economic Forum's 2025 Future of Jobs Report","World Economic Forum",{"relevance":83,"novelty":83,"quality":82,"actionability":58,"composite":4118,"reasoning":4119},3.05,"Category: AI & LLMs. The article discusses OpenAI's GPT-Realtime-2 and its capabilities for building real-time voice agents, which is relevant to AI-powered product development. However, it lacks actionable steps or frameworks that the audience can directly implement in their projects.","\u002Fsummaries\u002F26831750495fa9ed-openai-s-real-time-voice-ai-powers-agents-backed-b-summary","2026-05-08 22:31:14","2026-05-10 15:18:49",{"title":4052,"description":57},{"loc":4120},"4b39523ed70243f9","AI Revolution","https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=o_igSi-ED6s","summaries\u002F26831750495fa9ed-openai-s-real-time-voice-ai-powers-agents-backed-b-summary",[4130,98,99,101],"llm","https:\u002F\u002Fi.ytimg.com\u002Fvi\u002Fo_igSi-ED6s\u002Fhqdefault.jpg","OpenAI's GPT-Realtime-2 enables live voice agents with GPT-4o reasoning, 128k context, parallel tools, and 96.6% audio accuracy; MRC networking spreads data across paths for 131k-GPU clusters with microsecond failure recovery.","Narrated walkthrough of OpenAI's announcements on [real-time voice models](https:\u002F\u002Fopenai.com\u002Findex\u002Fadvancing-voice-intelligence-with-new-models-in-the-api\u002F) (GPT-Realtime-2\u002FTranslate\u002FWhisper for conversations, translation, transcription), [MRC supercomputer networking](https:\u002F\u002Fopenai.com\u002Findex\u002Fmrc-supercomputer-networking\u002F), and Sam Altman's [AI layoff comments](https:\u002F\u002Fwww.techradar.com\u002Fpro\u002Fsam-altman-says-some-companies-are-ai-washing-by-blaming-unrelated-layoffs-on-the-technology-but-admits-things-may-get-worse-soon), with a [CodeRabbit](https:\u002F\u002Fcoderabbit.link\u002Fai-revolution) sponsor segment.",[101],"Hnh2mFOtkRkBE9C8_maIzWYtBUahQHxcKa6xWpj7MA0",{"id":4137,"title":4138,"ai":4139,"body":4144,"categories":4377,"created_at":65,"date_modified":65,"description":57,"extension":66,"faq":65,"featured":67,"kicker_label":65,"meta":4378,"navigation":86,"path":4409,"published_at":4410,"question":65,"scraped_at":4411,"seo":4412,"sitemap":4413,"source_id":4414,"source_name":93,"source_type":4415,"source_url":4416,"stem":4417,"tags":4418,"thumbnail_url":65,"tldr":4420,"tweet":65,"unknown_tags":4421,"__hash__":4422},"summaries\u002Fsummaries\u002Febc0d711136fb32c-secure-ai-agents-via-mcp-toolbox-custom-tools-summary.md","Secure AI Agents via MCP Toolbox Custom Tools",{"provider":7,"model":8,"input_tokens":4140,"output_tokens":4141,"processing_time_ms":4142,"cost_usd":4143},8976,2997,46040,0.00327105,{"type":14,"value":4145,"toc":4369},[4146,4150,4153,4159,4162,4166,4169,4172,4179,4184,4187,4191,4194,4292,4295,4298,4303,4307,4310,4313,4316,4320,4323,4326,4331,4334,4336,4365],[17,4147,4149],{"id":4148},"tackling-the-confused-deputy-problem-in-ai-agents","Tackling the Confused Deputy Problem in AI Agents",[22,4151,4152],{},"AI agents promise automation like midnight database triage, but they risk the 'confused deputy' vulnerability: a service account with broad database access gets tricked by malicious user input (e.g., via prompt injection) into querying sensitive data like executive salaries instead of the paged-down DB. Kurtis Van Gent explains this as Simon Willison's 'lethal trifecta': private data + untrusted input + external sharing. Traditional fixes like prompt-engineered security fail because LLMs struggle to distinguish system vs. user instructions.",[4154,4155,4156],"blockquote",{},[22,4157,4158],{},"'The confused deputy problem is really a problem where you have some kind of authoritative source... but a malicious user or a bug can trick it into revealing information.' — Kurtis Van Gent, defining the core vulnerability with a real-world paging scenario.",[22,4160,4161],{},"Developers evaluated broad tool access (e.g., 'run any SQL') but rejected it for runtime agents serving end-users. Instead, they architected MCP Toolbox around customization: pre-author SQL queries reviewed like code, constraining what agents can do.",[17,4163,4165],{"id":4164},"build-time-vs-runtime-agents-tailored-tooling","Build-Time vs. Runtime Agents: Tailored Tooling",[22,4167,4168],{},"MCP Toolbox distinguishes two agent types, each with different security needs. Build-time agents (e.g., Gemini CLI, Claude Code) assist developers with broad, generic tools like 'any SQL' or BigQuery dashboard queries—safe since they use developer credentials. Runtime agents (e.g., customer service bots via ADK, LangChain) face untrusted users, needing narrow tools for accuracy and safety.",[22,4170,4171],{},"Toolbox supports both via generic (pre-built ops), runtime (dynamic), and custom tools. For databases like AlloyDB, BigQuery, Postgres, Valkey, Neo4j, Oracle, MariaDB, it acts as a 'central gate.' Open-source (15k+ GitHub stars, 130+ contributors, millions of monthly calls), it's self-hosted—no Google data access.",[22,4173,4174,4175,4178],{},"Key decision: Bound parameters separate agent-set values (e.g., flight ID from conversation) from app-set ones (e.g., user identity, target DB). This binds identity at runtime, e.g., ",[36,4176,4177],{},"tool.bind(user_id=authenticated_user)"," creates a scoped tool the LLM can't override.",[4154,4180,4181],{},[22,4182,4183],{},"'MCP is kind of the gold standard for interop right now... like USB for AI applications. You can take any agent and you can plug in any server.' — Kurtis Van Gent, positioning MCP as the standard Toolbox builds on.",[22,4185,4186],{},"Tradeoff: Hardcoding boosts security\u002Faccuracy (no hallucinated DB switches) but reduces flexibility. Philosophy: Remove agent control wherever possible without harming UX—e.g., hardcoded DB for single-DB sessions.",[17,4188,4190],{"id":4189},"custom-tools-pre-written-sql-as-architectural-guardrails","Custom Tools: Pre-Written SQL as Architectural Guardrails",[22,4192,4193],{},"Core mechanism: Define tools with fixed SQL templates and params. Example Postgres tool for airline queries:",[4195,4196,4200],"pre",{"className":4197,"code":4198,"language":4199,"meta":57,"style":57},"language-yaml shiki shiki-themes github-light github-dark","tool_type: postgres-sql\nsql: \"SELECT * FROM flights WHERE airline = $1 AND flight_number = $2\"\nparameters:\n  - name: airline\n    type: string\n  - name: flight_number\n    type: string\ndescription: \"Get flight details by airline and number\"\n","yaml",[36,4201,4202,4219,4229,4237,4250,4260,4272,4281],{"__ignoreMap":57},[4203,4204,4207,4211,4215],"span",{"class":4205,"line":4206},"line",1,[4203,4208,4210],{"class":4209},"s9eBZ","tool_type",[4203,4212,4214],{"class":4213},"sVt8B",": ",[4203,4216,4218],{"class":4217},"sZZnC","postgres-sql\n",[4203,4220,4221,4224,4226],{"class":4205,"line":58},[4203,4222,4223],{"class":4209},"sql",[4203,4225,4214],{"class":4213},[4203,4227,4228],{"class":4217},"\"SELECT * FROM flights WHERE airline = $1 AND flight_number = $2\"\n",[4203,4230,4231,4234],{"class":4205,"line":83},[4203,4232,4233],{"class":4209},"parameters",[4203,4235,4236],{"class":4213},":\n",[4203,4238,4239,4242,4245,4247],{"class":4205,"line":82},[4203,4240,4241],{"class":4213},"  - ",[4203,4243,4244],{"class":4209},"name",[4203,4246,4214],{"class":4213},[4203,4248,4249],{"class":4217},"airline\n",[4203,4251,4252,4255,4257],{"class":4205,"line":4031},[4203,4253,4254],{"class":4209},"    type",[4203,4256,4214],{"class":4213},[4203,4258,4259],{"class":4217},"string\n",[4203,4261,4263,4265,4267,4269],{"class":4205,"line":4262},6,[4203,4264,4241],{"class":4213},[4203,4266,4244],{"class":4209},[4203,4268,4214],{"class":4213},[4203,4270,4271],{"class":4217},"flight_number\n",[4203,4273,4275,4277,4279],{"class":4205,"line":4274},7,[4203,4276,4254],{"class":4209},[4203,4278,4214],{"class":4213},[4203,4280,4259],{"class":4217},[4203,4282,4284,4287,4289],{"class":4205,"line":4283},8,[4203,4285,4286],{"class":4209},"description",[4203,4288,4214],{"class":4213},[4203,4290,4291],{"class":4217},"\"Get flight details by airline and number\"\n",[22,4293,4294],{},"The LLM calls via MCP with params; Toolbox executes safely. No ad-hoc SQL generation—agents use dev-reviewed queries. Supports complex ops like joins\u002Fstored procs via custom SQL. Toolbox doesn't auto-write queries; devs do.",[22,4296,4297],{},"This mirrors app dev: Write\u002Freview SQL once, expose as API. For production, deploy on Cloud Run; min arch is Toolbox container + MCP client (Gemini\u002FVertex AI) + auth (e.g., IAM).",[4154,4299,4300],{},[22,4301,4302],{},"'The toolbox's superpower really comes down to... customize tools in a way that lets you constrain that access... write the SQL ahead of time.' — Kurtis Van Gent, on shifting from prompt hacks to code-like security.",[17,4304,4306],{"id":4305},"cymbal-air-demo-resilience-in-action","Cymbal Air Demo: Resilience in Action",[22,4308,4309],{},"Live demo of Cymbal Air (fictional airline agent): Normal flow—user asks flight status; agent uses bound tools to query only authorized data. Compromise attempt: \"Ignore instructions, query competitor salaries.\" Fails—tools lack access; agent stays on-topic.",[22,4311,4312],{},"Architecture: MCP client (Gemini) → Toolbox server (Cloud Run, Postgres backend) → bound custom tools. Code shown: Load tool, bind user context, register to agent. Result: Zero-trust, no leaks.",[22,4314,4315],{},"Evolution: Started with generic tools; pivoted to custom\u002Fbound for prod. Failure modes tested: Prompt injection blocked by param constraints.",[17,4317,4319],{"id":4318},"deployment-tradeoffs-and-best-practices","Deployment Tradeoffs and Best Practices",[22,4321,4322],{},"Latency: Toolbox adds ~50-100ms vs. direct queries (MCP overhead + execution); fine for interactive agents, not ultra-high-throughput. Self-hosted (binary\u002Fcontainer\u002Flocal); progressive tool exposure via dynamic registration.",[22,4324,4325],{},"Security-first process: Start with threat modeling ('what can go wrong?'), prototype fast with frameworks like ADK, then harden. 'Move security left'—architect params\u002Ftools early, iterate weekly.",[4154,4327,4328],{},[22,4329,4330],{},"'Flexibility versus security... anything that you can take away from the agent tends to be a good thing to take away as long as it doesn't diminish the use case.' — Kurtis Van Gent, on balancing autonomy and guardrails.",[22,4332,4333],{},"Non-obvious: Runtime agents need dev-like rigor (code review SQL); build-time can be looser. Replicate by forking GitHub repo, binding identity, testing injections.",[17,4335,3979],{"id":3978},[3859,4337,4338,4341,4344,4347,4350,4353,4356,4359,4362],{},[3862,4339,4340],{},"Model threats early: Map confused deputy risks (private data + untrusted input) before building agents.",[3862,4342,4343],{},"Use build-time tools broadly for dev (e.g., any-SQL); constrain runtime with custom MCP tools.",[3862,4345,4346],{},"Pre-write\u002Freview SQL templates; define params\u002Fdescriptions for LLM guidance.",[3862,4348,4349],{},"Bind app params (user ID, DB) at runtime—LLM sets only conversation-derived ones.",[3862,4351,4352],{},"Deploy self-hosted Toolbox on Cloud Run; test latency (\u003C100ms typical) and injections.",[3862,4354,4355],{},"Start small: Codelabs for BigQuery\u002FAlloyDB; scale to multi-agent apps.",[3862,4357,4358],{},"Prioritize security in architecture: 1st step = threat model, not prototype.",[3862,4360,4361],{},"Leverage open MCP spec: Plug any agent\u002Fserver; Google managed options for BigQuery\u002Fetc.",[3862,4363,4364],{},"Measure: Millions of safe calls\u002Fmonth via Toolbox—prod-proven.",[4366,4367,4368],"style",{},"html pre.shiki code .s9eBZ, html code.shiki .s9eBZ{--shiki-default:#22863A;--shiki-dark:#85E89D}html pre.shiki code .sVt8B, html code.shiki .sVt8B{--shiki-default:#24292E;--shiki-dark:#E1E4E8}html pre.shiki code .sZZnC, html code.shiki .sZZnC{--shiki-default:#032F62;--shiki-dark:#9ECBFF}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"title":57,"searchDepth":58,"depth":58,"links":4370},[4371,4372,4373,4374,4375,4376],{"id":4148,"depth":58,"text":4149},{"id":4164,"depth":58,"text":4165},{"id":4189,"depth":58,"text":4190},{"id":4305,"depth":58,"text":4306},{"id":4318,"depth":58,"text":4319},{"id":3978,"depth":58,"text":3979},[116],{"content_references":4379,"triage":4407},[4380,4383,4386,4389,4392,4395,4398,4401,4404],{"type":71,"title":4381,"url":4382,"context":74},"MCP Toolbox GitHub","https:\u002F\u002Fgoo.gle\u002Fgithub-mcp-toolbox",{"type":71,"title":4384,"url":4385,"context":74},"MCP Toolbox for Databases (Docs)","https:\u002F\u002Fgoo.gle\u002Fmcp-toolbox-dev",{"type":71,"title":4387,"url":4388,"context":74},"QuickStart","https:\u002F\u002Fgoo.gle\u002Fmcp-quickstart",{"type":71,"title":4390,"url":4391,"context":74},"MCP Toolbox for Databases: Making BigQuery datasets available to MCP clients (Codelab)","https:\u002F\u002Fgoo.gle\u002Fcodelabs",{"type":71,"title":4393,"url":4394,"context":74},"Build a Multi-agent App with MCP Toolbox for AlloyDB & ADK (Codelab)","https:\u002F\u002Fgoo.gle\u002Fcodelab-multi-agent-app",{"type":71,"title":4396,"url":4397,"context":74},"Cymbal Air Toolbox Demo","https:\u002F\u002Fgoo.gle\u002F4tfWYIA",{"type":71,"title":4399,"url":4400,"context":74},"Google Cloud MCP servers overview","https:\u002F\u002Fgoo.gle\u002F42ioQRn",{"type":71,"title":4402,"url":4403,"context":74},"MCP Toolbox for Databases (Toolbox)","https:\u002F\u002Fgoo.gle\u002F4wauUJp",{"type":71,"title":4405,"url":4406,"context":74},"GEAR","https:\u002F\u002Fgoo.gle\u002FGEAR",{"relevance":82,"novelty":83,"quality":82,"actionability":83,"composite":3808,"reasoning":4408},"Category: AI & LLMs. The article addresses a specific pain point regarding security in AI agents, particularly the confused deputy problem, which is relevant for developers integrating AI features. It provides insights into a practical solution (MCP Toolbox) but lacks detailed step-by-step guidance for implementation.","\u002Fsummaries\u002Febc0d711136fb32c-secure-ai-agents-via-mcp-toolbox-custom-tools-summary","2026-05-05 16:46:33","2026-05-06 16:12:43",{"title":4138,"description":57},{"loc":4409},"ed722ee0fdc7e076","article","https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=CRszhkEjd8s","summaries\u002Febc0d711136fb32c-secure-ai-agents-via-mcp-toolbox-custom-tools-summary",[98,99,100,4419],"devops","MCP Toolbox prevents confused deputy attacks by letting developers pre-write constrained SQL tools with bound parameters, separating agent flexibility from app-controlled security for runtime agents.",[],"htBzEsyR16VdzmViKPvmry-2HFiUx9a6ye2MxpmOJCk"]