1~
2Claude Agent SDK daiot programmnyi dostup k tomu zhe tsiklu agenta, kotoryi stoit za Claude Code. Vashi agenty mogut chitat faily, vypolniat komandy obolochki, iskat v internete, redaktirovat kod, vyzyvat vneshnie API cherez MCP-servery i orkestirovat sub-agentov  -  vsio eto iz neskolkikh strok TypeScript ili Python.
3~
4V otlichie ot standartnogo Anthropic Client SDK, gde vy sami stroite tsikl instrumentov, Agent SDK vnutrenne obrabatyvaet vypolnenie instrumentov, upravlenie kontekstom, povtornye popytki i orkestratsiiu. Vy opisyvaete to, chto khotite, predostavliaete instrumenty, a agent razbiraitsia s ostalnym.
5~
6## Arkhitektura
7~
8SDK sleduet prostomu tsiklu: **soberi kontekst, vypolni deistvie, prover, povtori**.
9~
10```mermaid
11graph TD
12    Input[User Prompt] --> Loop[Agent Loop]
13    Loop --> Reason[Claude Reasons]
14    Reason --> Tool[Call Tool]
15    Tool --> Result[Tool Result]
16    Result --> Loop
17    Reason --> Done[Final Response]
18~
19    subgraph "Built-in Tools"
20        T1[Read / Write / Edit]
21        T2[Bash / Terminal]
22        T3[Glob / Grep]
23        T4[WebSearch / WebFetch]
24    end
25~
26    subgraph "Custom Tools via MCP"
27        M1[Your API]
28        M2[Database]
29        M3[Slack / GitHub / etc.]
30    end
31~
32    Tool --> T1
33    Tool --> M1
34```
35~
36Osnovnaia tochka vkhoda  -  `query()`, kotoraia vozvrashchaet asinkhronny iterator, transliriuiushchii soobshcheniia po mere raboty agenta. Kazhdoe soobshchenie govori vam, chto delaet agent: razmyshliaet, vyzyvaet instrument, poluchaet rezultat ili dostavliaet itogovyi vyvod.
37~
38## Nachalo raboty
39~
40### Ustanovka
41~
42```bash
43# TypeScript
44npm install @anthropic-ai/claude-agent-sdk
45~
46# Python
47pip install claude-agent-sdk
48```
49~
50Vam nuzhen API-kliuch Anthropic, ustanovlenny kak `ANTHROPIC_API_KEY` v vashem okruzhenii.
51~
52### Vash pervyi agent
53~
54```typescript
55import { query } from "@anthropic-ai/claude-agent-sdk";
56~
57const conversation = query({
58  prompt: "Find all TODO comments in the codebase and create a summary",
59  options: {
60    allowedTools: ["Read", "Glob", "Grep"],
61  },
62});
63~
64for await (const message of conversation) {
65  if (message.type === "assistant") {
66    process.stdout.write(message.content);
67  }
68  if (message.type === "result" && message.subtype === "success") {
69    console.log("\nDone:", message.result);
70  }
71}
72```
73~
74Vot i vsio. Agent ispolzuet Glob dlia poiska failov, Grep dlia poiska patternov TODO, Read dlia izucheniia sovpadenii i verniot strukturirovannoe reziume. Vy ne pishete logiku orkestratsii  -  SDK berot eto na sebia.
75~
76### Ekvivalent na Python
77~
78```python
79from claude_agent_sdk import query
80~
81async for message in query(
82    prompt="Find all TODO comments in the codebase and create a summary",
83    options={"allowed_tools": ["Read", "Glob", "Grep"]},
84):
85    if message.type == "assistant":
86        print(message.content, end="")
87    if message.type == "result" and message.subtype == "success":
88        print(f"\nDone: {message.result}")
89```
90~
91## Vstroennye instrumenty
92~
93SDK postavliaetsia s temi zhe instrumentami, kotorye dostupny v Claude Code:
94~
95| Instrument | Opisanie |
96|------|-------------|
97| **Read** | Chtenie soderzhimogo failov |
98| **Write** | Sozdanie novykh failov |
99| **Edit** | Tselevye pravki v sushchestvuiushchikh failakh |
100| **Bash** | Vypolnenie komand obolochki |
101| **Glob** | Poisk failov po patternu |
102| **Grep** | Poisk v soderzhimom failov s pomoshchiu regex |
103| **WebSearch** | Poisk v internete |
104| **WebFetch** | Zagruzka URL i vozvrat ego soderzhimogo |
105| **AskUserQuestion** | Zapros vvoda ot polzovatelia |
106~
107Vy kontroliruete, kakie instrumenty agent mozhet ispolzovat, cherez `allowedTools`. Esli instrumenta net v spiske, agent ne mozhet ego vyzvat.
108~
109## Rezhimy razreshenii
110~
111Poskolku agenty vypolniaiut nastoyashchie komandy na nastoyashchikh sistemakh, razresheniia vazhny.
112~
113| Rezhim | Povedenie | Primenenie |
114|------|----------|----------|
115| `default` | Polzovatelsky callback `canUseTool` reshaet pri kazhdom vyzove | Tonkaia nastroika |
116| `acceptEdits` | Avtoodobrenie failovykh operatsii, zapros dlia Bash | Rabochie protsessy razrabotki |
117| `dontAsk` | Otkaz vo vsem, chto ne v allowedTools | Ogranichennye agenty |
118| `bypassPermissions` | Avtomaticheskoe odobrenie vsego | Doverennye sandbox-sredy |
119| `auto` | Klassifikator modeli reshaet o bezopasnosti | Sbalansirovannaia avtomatizatsiia |
120~
121```typescript
122const conversation = query({
123  prompt: "Refactor the auth module to use JWT",
124  options: {
125    allowedTools: ["Read", "Edit", "Glob", "Grep", "Bash"],
126    permissionMode: "acceptEdits",
127  },
128});
129```
130~
131Dlia produktsionnogo ispolzovaniia vsegda zapuskaite agentov v izolirovannykh sredakh (konteinery, virtualnye mashiny) i ispolzuyte samyi strogiiy rezhim razreshenii, kotoryi vsio eshchio pozvoliaet agentu vypolniat svoiu rabotu.
132~
133## Sozdanie polzovatelskikh instrumentov s MCP
134~
135Nastoiashchaia sila SDK zakliuchaetsia v rasshirenii agentov vashimi sobstvennymi instrumentami. Polzovatelskie instrumenty opredeliaiutsia kak in-process MCP-servery  -  bez upravleniia podprotsessami, bez setevykh nakladnykh raskhodov.
136~
137### Primer: Instrument pogody
138~
139```typescript
140import { tool, createSdkMcpServer, query } from "@anthropic-ai/claude-agent-sdk";
141import { z } from "zod";
142~
143const getTemperature = tool(
144  "get_temperature",
145  "Get the current temperature at a location",
146  {
147    latitude: z.number().describe("Latitude"),
148    longitude: z.number().describe("Longitude"),
149  },
150  async ({ latitude, longitude }) => {
151    const res = await fetch(
152      `https://api.open-meteo.com/v1/forecast?latitude=${latitude}&longitude=${longitude}&current=temperature_2m&temperature_unit=celsius`
153    );
154    const data = await res.json();
155    return {
156      content: [
157        {
158          type: "text",
159          text: `Current temperature: ${data.current.temperature_2m}C`,
160        },
161      ],
162    };
163  }
164);
165~
166const weatherServer = createSdkMcpServer({
167  name: "weather",
168  version: "1.0.0",
169  tools: [getTemperature],
170});
171~
172for await (const message of query({
173  prompt: "What's the weather like in Rome?",
174  options: {
175    mcpServers: { weather: weatherServer },
176    allowedTools: ["mcp__weather__get_temperature"],
177  },
178})) {
179  if (message.type === "result" && message.subtype === "success") {
180    console.log(message.result);
181  }
182}
183```
184~
185Polzovatelskie instrumenty sleduiut konventsii imenovaniia `mcp__{imia_servera}__{imia_instrumenta}`. Vy mozhete ispolzovat podstanovochnye znaki v `allowedTools`: `"mcp__weather__*"` razreshaet vse instrumenty s servera pogody.
186~
187### Primer: Instrument zaprosov k baze dannykh
188~
189```typescript
190const queryDb = tool(
191  "query_database",
192  "Run a read-only SQL query against the application database",
193  {
194    sql: z.string().describe("SQL SELECT query to execute"),
195  },
196  async ({ sql }) => {
197    // Validate: only allow SELECT queries
198    if (!sql.trim().toUpperCase().startsWith("SELECT")) {
199      return {
200        content: [{ type: "text", text: "Error: Only SELECT queries are allowed." }],
201      };
202    }
203~
204    const result = await pool.query(sql);
205    return {
206      content: [
207        {
208          type: "text",
209          text: JSON.stringify(result.rows, null, 2),
210        },
211      ],
212    };
213  }
214);
215```
216~
217## Podkliuchenie vneshnikh MCP-serverov
218~
219Pomimo in-process instrumentov, vy mozhete podkliuchatsia k liubomu sushchestvuiushchemu MCP-serveru  -  tem zhe serveram, kotorye rabotaiut s Claude Desktop, Cursor i drugimi MCP-klientami.
220~
221```typescript
222for await (const message of query({
223  prompt: "Check the latest issues in the frontend repo and summarize them",
224  options: {
225    mcpServers: {
226      github: {
227        command: "npx",
228        args: ["-y", "@modelcontextprotocol/server-github"],
229        env: { GITHUB_PERSONAL_ACCESS_TOKEN: process.env.GITHUB_TOKEN },
230      },
231    },
232    allowedTools: ["mcp__github__*"],
233  },
234})) {
235  // ...
236}
237```
238~
239Vy mozhete kombinirovat neskolko MCP-serverov. Agent vidit vse instrumenty so vsekh podkliuchennykh serverov i ispolzuet ikh po mere neobkhodimosti.
240~
241```mermaid
242graph LR
243    Agent[Your Agent] --> SDK[Agent SDK]
244    SDK --> InProcess[In-process MCP\nCustom Tools]
245    SDK --> GitHub[GitHub MCP Server]
246    SDK --> Postgres[PostgreSQL MCP Server]
247    SDK --> Slack[Slack MCP Server]
248```
249~
250## Orkestratsiia neskolkikh agentov
251~
252Dlia slozhnykh rabochikh protsessov vy mozhete opredelit spetsializirovannykh sub-agentov, kotorym roditelsky agent delegiruet zadachi. U kazhdogo sub-agenta svoi prompt, instrumenty i oblast fokusa.
253~
254```typescript
255for await (const message of query({
256  prompt: "Review the PR, check for security issues, and update the changelog",
257  options: {
258    allowedTools: ["Read", "Edit", "Bash", "Glob", "Grep", "Agent"],
259    agents: [
260      {
261        name: "security-reviewer",
262        description: "Reviews code for security vulnerabilities",
263        prompt: "You are a security expert. Analyze code for OWASP Top 10 vulnerabilities.",
264        allowedTools: ["Read", "Glob", "Grep"],
265      },
266      {
267        name: "changelog-writer",
268        description: "Updates the CHANGELOG.md file based on recent changes",
269        prompt: "You maintain the project changelog. Follow Keep a Changelog format.",
270        allowedTools: ["Read", "Edit", "Bash"],
271      },
272    ],
273  },
274})) {
275  // The parent agent will:
276  // 1. Read the PR diff
277  // 2. Delegate security review to security-reviewer
278  // 3. Delegate changelog update to changelog-writer
279  // 4. Synthesize results
280}
281```
282~
283Dobavte `"Agent"` v `allowedTools` roditelskogo agenta dlia vkliucheniia delegirovaniia. Sub-agenty rabotaiut so svoimi instrumentami i ne imeiut dostupa k instrumentam roditelia, esli eto ne predostavleno iavno.
284~
285```mermaid
286graph TD
287    Parent[Parent Agent] --> SR[Security Reviewer\nRead, Glob, Grep]
288    Parent --> CW[Changelog Writer\nRead, Edit, Bash]
289    SR --> Report[Security Report]
290    CW --> Updated[Updated CHANGELOG]
291    Report --> Parent
292    Updated --> Parent
293    Parent --> Final[Final Summary]
294```
295~
296## Sessii i nepreryvnost
297~
298Agenty mogut sokhraniat kontekst mezhdu neskolkimi zaprosami s pomoshchiu sessii. Zakhvatite `session_id` iz pervogo vzaimodeistviia i peredaite ego v `resume` dlia poslesuiushchikh zaprosov.
299~
300```typescript
301let sessionId: string | undefined;
302~
303// First query
304for await (const message of query({
305  prompt: "Read the project structure and understand the architecture",
306  options: { allowedTools: ["Read", "Glob", "Grep"] },
307})) {
308  if (message.type === "init") {
309    sessionId = message.session_id;
310  }
311}
312~
313// Follow-up query (same session, full context preserved)
314for await (const message of query({
315  prompt: "Now refactor the auth module based on what you learned",
316  resume: sessionId,
317  options: { allowedTools: ["Read", "Edit", "Bash"] },
318})) {
319  // Agent remembers the full project context from the first query
320}
321```
322~
323## Claude Managed Agents
324~
325Esli vy ne khotite razvorachivat infrastrukturu agenta samostoiatelno, **Claude Managed Agents** (zapushcheny v aprele 2026) predostavliaet polnostiu upravliaemyi oblachny servis. Anthropic zapuskaet konteinery, obrabatyvaet masshtabirovanie i predostavliaet streaming API.
326~
327```mermaid
328graph LR
329    subgraph "Self-hosted (Agent SDK)"
330        Code[Your Code] --> SDK2[Agent SDK]
331        SDK2 --> API[Claude API]
332        SDK2 --> Tools[Your Tools]
333    end
334~
335    subgraph "Managed Agents"
336        App[Your App] --> MAPI[Managed Agents API]
337        MAPI --> Container[Anthropic-hosted Container]
338        Container --> API2[Claude API]
339        Container --> Tools2[Tools in Container]
340    end
341```
342~
343Kliuchevoe razlichie: s Agent SDK vy zapuskaete tsikl agenta v svoei sobstvennoi infrastrukture. S Managed Agents Anthropic razmeshshaet i zapuskaet agenta za vas. Vy vzaimodeistuvete cherez API na osnove sessii i poluchaete sobytiia cherez Server-Sent Events.
344~
345**Tseny:**
346- **Agent SDK**: tolko standartnye tarify na tokeny Claude API. Khosting na vashei storone.
347- **Managed Agents**: tarify na tokeny plius 0,08 USD za chas sessii (tarifikatsiia pomillisekundno).
348~
349## Luchshie praktiki dlia produktsii
350~
351### 1. Vsegda izoliruyte
352~
353Nikogda ne zapuskaite agentov s neogranichennymi razresheniiami na produktsionnoi mashine. Ispolzuite konteinery (Docker, Fly.io, Modal) ili izolirovannye sredy (E2B, Vercel Sandbox).
354~
355### 2. Ogranichivaite dostup k instrumentam
356~
357Sleduyte printsipu naimenshikh privilegii. Agentu, generiruiushchemu otchioty, ne nuzhny `Bash` ili `Write`.
358~
359```typescript
360// Too permissive
361allowedTools: ["Read", "Write", "Edit", "Bash", "Glob", "Grep"]
362~
363// Better: only what's needed
364allowedTools: ["Read", "Glob", "Grep"]
365```
366~
367### 3. Ispolzuyte hooks dlia zashchitnykh ograzhdenii
368~
369Hooks pozvoliaiut perekhvatyvat vyzovy instrumentov do i posle vypolneniia. Ispolzuyte ikh dlia logirovaniia, validatsii i ogranicheniia chastoty.
370~
371```typescript
372const conversation = query({
373  prompt: "Analyze the codebase",
374  options: {
375    allowedTools: ["Read", "Glob", "Grep"],
376    hooks: {
377      PreToolUse: async (toolName, input) => {
378        console.log(`Tool call: ${toolName}`, input);
379        // Return false to block the call
380        if (toolName === "Bash" && input.command.includes("rm")) {
381          return false;
382        }
383        return true;
384      },
385    },
386  },
387});
388```
389~
390### 4. Obrabatyvyte oshibki akkuratno
391~
392Tsikl agenta mozhet generirovat oshibki  -  sboi instrumentov, limity API, perepolnenie kontekstnogo okna. Vsegda proveryaite tipy soobshchenii.
393~
394```typescript
395for await (const message of conversation) {
396  switch (message.type) {
397    case "assistant":
398      // Agent reasoning
399      break;
400    case "tool_use":
401      // Agent is calling a tool
402      break;
403    case "result":
404      if (message.subtype === "error") {
405        console.error("Agent failed:", message.error);
406      }
407      break;
408  }
409}
410```
411~
412### 5. Monitorte ispolzovanie tokenov
413~
414Tsikly agentov mogut potrebliat znachitelnoe kolichestvo tokenov, osobenno s bolshimi bazami koda. SDK vkliuchaet avtomaticheskuiu kompaktsiiu konteksta, no vy vsio ravno dolzhny monitorit ispolzovanie.
415~
416## Zakliuchenie
417~
418Claude Agent SDK prevrashchaet LLM iz mashiny dlia otvetov na voprosy v chto-to blizhe k junior-razrabotchiku. Vashi agenty mogut chitat, pisat, vypolniat, proveriat i iteirovat  -  tot zhe workflow, kotoromu sleduet chelovek.
419~
420Nachinte s malogo: sozdaite agenta s neskolkimi vstroennymi instrumentami. Zatem dobavte polzovatelskie MCP-instrumenty dlia vashei konkretnoi oblasti. Masshtabiruyte do orkestratsii neskolkikh agentov, kogda vashi protsessy trebuiut spetsializatsii.
421~
422Tsikl agenta  -  tot zhe, kotoryi stoit za Claude Code. Esli on mozhet sozdavat programmnoe obespechenie, vashi agenty toge smogut.
423~
424> **Chek-list dlia nachala:**
425>
426> - [x] Ustanovite SDK (`npm install @anthropic-ai/claude-agent-sdk`)
427> - [x] Ustanovite `ANTHROPIC_API_KEY` v vashem okruzhenii
428> - [x] Sozdaite prostogo agenta so vstroennymi instrumentami (Read, Glob, Grep)
429> - [x] Dobavte polzovatelsky instrument cherez in-process MCP-server
430> - [x] Podkliuchite vneshni MCP-server (GitHub, PostgreSQL, i t.d.)
431> - [x] Realizuyte orkestratsiiu neskolkikh agentov s sub-agentami
432> - [x] Nastroite izolirovannuiu sredu dlia produktsii
433> - [x] Dobavte hooks dlia logirovaniia i zashchitnykh ograzhdenii
434~