@dmsdc-ai/aigentry-deliberation 0.0.37 → 0.0.38

package/README.md

@@ -33,7 +33,7 @@ MCP Deliberation Server — Multi-session AI deliberation with smart speaker ord
 원클릭 설치 — 어떤 프로젝트 환경에서든 동작합니다:
 
 ```bash
-npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install
+npx @dmsdc-ai/aigentry-deliberation install
 ```
 
 이 명령은:
@@ -62,7 +62,7 @@ cd aigentry-deliberation && npm install && node install.js
 ### 제거
 
 ```bash
-npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install --uninstall
+npx @dmsdc-ai/aigentry-deliberation uninstall
 ```
 
 MCP 서버 등록 해제 + 설치 파일 삭제 + 스킬 파일 정리까지 자동 처리됩니다.
@@ -87,7 +87,7 @@ open demo/forum/index.html
 MCP 연결 문제 자동 진단:
 
 ```bash
-npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-doctor
+npx @dmsdc-ai/aigentry-deliberation doctor
 ```
 
 Claude Code, Codex CLI, Gemini CLI의 MCP 설정을 자동 점검하고 문제를 진단합니다.
@@ -101,7 +101,6 @@ Claude Code, Codex CLI, Gemini CLI의 MCP 설정을 자동 점검하고 문제
 | `deliberation_synthesize` | Generate synthesis report |
 | `deliberation_status` | Check session status |
 | `deliberation_context` | Load project context |
-| `deliberation_inject_context` | Inject structured context or external experiment history into an active session |
 | `deliberation_history` | View discussion history |
 | `deliberation_list_active` | List active sessions |
 | `deliberation_list` | List archived sessions |
@@ -138,48 +137,6 @@ Telepty-managed sessions are now routed through the telepty bus instead of raw P
 - `session_health` bus events are cached for operator visibility
 - `deliberation_synthesize` validates and emits typed `deliberation_completed` envelopes for downstream automation
 
-## Experiment Retrospectives
-
-For autoresearch-style keep/discard reviews, inject a compact experiment bundle after the session starts instead of bloating `topic`.
-
-Recommended rules:
-- keep the injected JSON around `1.5KB` to `2KB`
-- include only the last `3-5` relevant experiments
-- keep `key_changes` to at most `3` scalar before/after pairs
-- reference bulky artifacts (`results.tsv`, full `program.md`, JSONL logs) by path only
-
-Example:
-
-```text
-deliberation_start(...)
-deliberation_inject_context(
-  session_id: "experiment-review-123",
-  speaker: "dustcraw",
-  context: "{\"past_experiments\":[{\"experiment_id\":\"dg-20260310-001\",\"signal_kind\":\"INTEREST_DRIFT\",\"patch_summary\":\"Raised relevanceThreshold from 0.30 to 0.35\",\"patch_kind\":\"config\",\"key_changes\":{\"relevanceThreshold\":{\"before\":0.3,\"after\":0.35}},\"score\":0.08,\"score_label\":\"promotion_rate_delta\",\"metric_name\":\"promotion_rate_delta\",\"metric_delta\":0.08,\"verdict\":\"positive\",\"followup_action\":\"kept\",\"reasoning\":\"Threshold raise reduced noise; promotion quality improved 8%\"}],\"experiment_count\":1,\"success_rate\":1.0}"
-)
-```
-
-If your synthesis needs an explicit experiment verdict, `structured` can now include `experiment_outcome`:
-
-```json
-{
-  "summary": "Lower the blast radius and re-run with stricter constraints.",
-  "decisions": [
-    "Keep the experiment loop bounded to one editable file",
-    "Retry after restoring the failing test baseline"
-  ],
-  "actionable_tasks": [
-    { "id": 1, "task": "Tighten editable globs", "project": "aigentry-devkit", "priority": "high" }
-  ],
-  "experiment_outcome": {
-    "verdict": "modify",
-    "suggested_action": "iterate",
-    "confidence": 0.78,
-    "measurement_window_hours": 24
-  }
-}
-```
-
 ## Speaker Ordering Strategies
 
 | Strategy | Description |
@@ -239,7 +196,7 @@ Inserts multi-AI verification gates at key [superpowers](https://github.com/obra
 
 **Fallback:** MCP 미설치 시 self-criticism 기반 자가 검증으로 대체 (Silver 등급). MCP 설치 시 멀티-AI 토론 (Gold 등급).
 
-**Install:** `npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install` 실행 시 자동 설치됩니다.
+**Install:** `npx @dmsdc-ai/aigentry-deliberation install` 실행 시 자동 설치됩니다.
 
 수동 설치:
 ```bash

package/doctor.js

@@ -15,7 +15,7 @@
  *
  * Usage:
  *   node doctor.js
- *   npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-doctor
+ *   npx @dmsdc-ai/aigentry-deliberation doctor
  */
 
 import fs from "node:fs";
@@ -209,7 +209,7 @@ function suggestFix(serverName, server, issue) {
   switch (issue) {
     case "path_missing":
       if (serverName === "deliberation" || serverName === "mcp-deliberation") {
-        return `npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install`;
+        return `npx @dmsdc-ai/aigentry-deliberation install`;
       }
       if (serverName.includes("brain") || serverName.includes("aigentry-brain")) {
         return `npx @dmsdc-ai/aigentry-brain install`;
@@ -223,7 +223,7 @@ function suggestFix(serverName, server, issue) {
     case "temp_path": {
       const tempArg = (server.args || []).find((a) => isTempPath(a));
       if (serverName === "deliberation" || serverName === "mcp-deliberation") {
-        return `npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install  # reinstall to permanent path`;
+        return `npx @dmsdc-ai/aigentry-deliberation install  # reinstall to permanent path`;
       }
       if (serverName.includes("brain") || serverName.includes("aigentry-brain")) {
         return `npx @dmsdc-ai/aigentry-brain install  # reinstall to permanent path`;
@@ -357,7 +357,7 @@ function runDiagnostics() {
         if (mod.includes("aigentry-brain") || mod.includes("brain")) {
           fix = `npx @dmsdc-ai/aigentry-brain install  # temporary path → reinstall to permanent path`;
         } else if (mod.includes("deliberation") || mod.includes("mcp-deliberation")) {
-          fix = `npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install  # temporary path -> reinstall to permanent path`;
+          fix = `npx @dmsdc-ai/aigentry-deliberation install  # temporary path → reinstall to permanent path`;
         } else {
           fix = `# temporary path (${mod}) — change to permanent path in MCP config`;
         }
@@ -387,7 +387,7 @@ function runDiagnostics() {
   } else {
     totalIssues++;
     console.log(`   ❌ Server file not found: ${selfPath}`);
-    console.log(`      fix: npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install`);
+    console.log(`      fix: npx @dmsdc-ai/aigentry-deliberation install`);
   }
 
   // Check node_modules
@@ -407,7 +407,7 @@ function runDiagnostics() {
   } catch {
     totalIssues++;
     console.log(`   ❌ Syntax error detected`);
-    console.log(`      fix: npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install`);
+    console.log(`      fix: npx @dmsdc-ai/aigentry-deliberation install`);
   }
 
   // ── Summary ──

package/index.js

@@ -22,10 +22,9 @@ if (_cliArg === "--help" || _cliArg === "-h") {
 MCP Deliberation Server
 
 Usage:
-  npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install     Install (preferred)
-  npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install --uninstall
+  npx @dmsdc-ai/aigentry-deliberation install     Install (register MCP server)
+  npx @dmsdc-ai/aigentry-deliberation uninstall    Uninstall
   npx @dmsdc-ai/aigentry-deliberation              Run MCP server (stdio)
-  npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-doctor      Diagnose MCP wiring
 
 After installation, restart Claude Code to start using it.
 `);
@@ -144,32 +143,10 @@ const StructuredActionableTaskSchema = z.object({
   priority: z.enum(["high", "medium", "low"]).optional(),
 });
 
-const StructuredExperimentOutcomeSchema = z.object({
-  verdict: z.enum(["keep", "discard", "modify"]),
-  confidence: z.number().min(0).max(1).optional(),
-  measurement_window_hours: z.number().nonnegative().optional(),
-  patches: z.array(z.unknown()).optional(),
-  suggested_action: z.enum(["advance", "revert", "iterate"]).optional(),
-});
-
 const StructuredSynthesisSchema = z.object({
   summary: z.string(),
   decisions: z.array(z.string()),
   actionable_tasks: z.array(StructuredActionableTaskSchema),
-  experiment_outcome: StructuredExperimentOutcomeSchema.optional(),
-});
-
-const StructuredExecutionContractSchema = z.object({
-  version: z.literal("v1"),
-  source_session_id: z.string().min(1),
-  summary: z.string(),
-  tasks: z.array(StructuredActionableTaskSchema),
-  experiment_outcome: StructuredExperimentOutcomeSchema.nullable().optional(),
-  unresolved_questions: z.array(z.string()),
-  artifact_refs: z.array(z.string()),
-  generated_from: z.object({
-    structured_synthesis_hash: z.string().length(40),
-  }),
 });
 
 const TeleptyEnvelopeSchema = z.object({
@@ -202,7 +179,6 @@ const TeleptyDeliberationCompletedPayloadSchema = z.object({
   topic: z.string(),
   synthesis: z.string(),
   structured_synthesis: StructuredSynthesisSchema.nullable().optional(),
-  execution_contract: StructuredExecutionContractSchema.nullable().optional(),
 });
 
 const TELEPTY_ENVELOPE_PAYLOAD_SCHEMAS = {
@@ -524,41 +500,6 @@ function hashPromptText(value) {
   return createHash("sha1").update(String(value || "")).digest("hex");
 }
 
-function sortJsonValue(value) {
-  if (Array.isArray(value)) {
-    return value.map(sortJsonValue);
-  }
-  if (value && typeof value === "object") {
-    return Object.keys(value)
-      .sort()
-      .reduce((acc, key) => {
-        acc[key] = sortJsonValue(value[key]);
-        return acc;
-      }, {});
-  }
-  return value;
-}
-
-function hashStructuredSynthesis(structured) {
-  return hashPromptText(JSON.stringify(sortJsonValue(structured || null)));
-}
-
-function buildExecutionContract({ state, structured }) {
-  if (!structured) return null;
-  return {
-    version: "v1",
-    source_session_id: state.id,
-    summary: structured.summary || "",
-    tasks: structured.actionable_tasks || [],
-    experiment_outcome: structured.experiment_outcome || null,
-    unresolved_questions: [],
-    artifact_refs: [],
-    generated_from: {
-      structured_synthesis_hash: hashStructuredSynthesis(structured),
-    },
-  };
-}
-
 function createEnvelopeId(prefix = "env") {
   return `${prefix}-${Date.now().toString(36)}-${Math.random().toString(36).slice(2, 10)}`;
 }
@@ -619,11 +560,7 @@ function buildTeleptyTurnRequestEnvelope({ state, speaker, turnId, turnPrompt, i
   });
 }
 
-function buildTeleptySynthesisEnvelope({ state, synthesis, structured, executionContract }) {
-  const derivedExecutionContract =
-    executionContract !== undefined
-      ? executionContract
-      : (structured ? buildExecutionContract({ state, structured }) : (state.execution_contract || null));
+function buildTeleptySynthesisEnvelope({ state, synthesis, structured }) {
   return buildTeleptyEnvelope({
     session_id: state.id,
     project: state.project || getProjectSlug(),
@@ -639,7 +576,6 @@ function buildTeleptySynthesisEnvelope({ state, synthesis, structured, execution
       topic: state.topic,
       synthesis,
       structured_synthesis: structured || null,
-      execution_contract: derivedExecutionContract || null,
     },
   });
 }
@@ -2611,14 +2547,6 @@ tags: [deliberation]
     md += `## Synthesis\n\n${s.synthesis}\n\n---\n\n`;
   }
 
-  if (s.structured_synthesis) {
-    md += `## Structured Synthesis\n\n\`\`\`json\n${JSON.stringify(s.structured_synthesis, null, 2)}\n\`\`\`\n\n---\n\n`;
-  }
-
-  if (s.execution_contract) {
-    md += `## Execution Contract\n\n\`\`\`json\n${JSON.stringify(s.execution_contract, null, 2)}\n\`\`\`\n\n---\n\n`;
-  }
-
   md += `## Debate Log\n\n`;
   for (const entry of s.log) {
     md += `### ${entry.speaker} — Round ${entry.round}\n\n`;
@@ -4542,7 +4470,6 @@ async function runAutoHandoff(sessionId) {
       if (!loaded) return;
       loaded.synthesis = markdownSynthesis;
       loaded.structured_synthesis = structured;
-      loaded.execution_contract = buildExecutionContract({ state: loaded, structured });
       loaded.status = "completed";
       loaded.current_speaker = "none";
       saveSession(loaded);
@@ -5085,7 +5012,6 @@ server.tool(
 
       loaded.synthesis = synthesis;
       loaded.structured_synthesis = structured || null;
-      loaded.execution_contract = buildExecutionContract({ state: loaded, structured: structured || null });
       loaded.status = "completed";
       loaded.current_speaker = "none";
       saveSession(loaded);
@@ -5107,7 +5033,6 @@ server.tool(
       state,
       synthesis,
       structured,
-      executionContract: state.execution_contract || null,
     });
 
     // Immediately force-close monitor terminal (including physical Terminal) on deliberation end

package/install.js

@@ -4,13 +4,13 @@
  * Deliberation MCP Server — One-click installer
  *
  * Usage:
- *   npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install
+ *   npx @dmsdc-ai/aigentry-deliberation install
  *   node install.js
  *
  * What it does:
  *   1. Copies server files to ~/.local/lib/mcp-deliberation/
  *   2. Installs npm dependencies
- *   3. Registers MCP server in ~/.claude/.mcp.json (Claude Code)
+ *   3. Registers MCP server via `claude mcp add` (Claude Code)
  *   4. Registers MCP server in ~/.gemini/settings.json (Gemini CLI)
  *   5. Ready to use — next Claude Code or Gemini CLI session will auto-load
  *   6. Installs skill file (~/.claude/skills/deliberation-gate/SKILL.md)
@@ -61,6 +61,15 @@ function log(msg) {
   console.log(`  ${msg}`);
 }
 
+function commandExists(cmd) {
+  try {
+    execSync(IS_WIN ? `where ${cmd}` : `command -v ${cmd}`, { stdio: "pipe" });
+    return true;
+  } catch {
+    return false;
+  }
+}
+
 function copyFileIfExists(src, dest) {
   if (fs.existsSync(src)) {
     fs.copyFileSync(src, dest);
@@ -121,32 +130,55 @@ function install() {
     log("   Manual fix: cd ~/.local/lib/mcp-deliberation && npm install");
   }
 
-  // Step 4: Register MCP server
+  // Step 4: Register MCP server (Claude Code)
   log("🔧 Registering Claude Code MCP server...");
-  const claudeDir = path.join(HOME, ".claude");
-  fs.mkdirSync(claudeDir, { recursive: true });
+  const serverEntryPoint = toForwardSlash(path.join(INSTALL_DIR, "index.js"));
+  let claudeRegistered = false;
 
-  let mcpConfig = {};
-  if (fs.existsSync(MCP_CONFIG)) {
+  // Prefer `claude mcp add` — this is the only method Claude Code reliably reads
+  if (commandExists("claude")) {
     try {
-      mcpConfig = JSON.parse(fs.readFileSync(MCP_CONFIG, "utf-8"));
-    } catch {
-      mcpConfig = {};
+      // Remove existing registration first (ignore errors if not registered)
+      try { execSync("claude mcp remove deliberation -s user", { stdio: "pipe" }); } catch { /* ok */ }
+      execSync(`claude mcp add deliberation -s user -- node "${serverEntryPoint}"`, {
+        stdio: "pipe",
+      });
+      log("   → Registered via 'claude mcp add' (user scope)");
+      claudeRegistered = true;
+    } catch (err) {
+      log(`   ⚠️ 'claude mcp add' failed: ${err.message}`);
     }
   }
 
-  if (!mcpConfig.mcpServers) mcpConfig.mcpServers = {};
+  // Fallback: write ~/.claude/.mcp.json (legacy, may not be read by Claude Code)
+  if (!claudeRegistered) {
+    const claudeDir = path.join(HOME, ".claude");
+    fs.mkdirSync(claudeDir, { recursive: true });
 
-  const alreadyRegistered = !!mcpConfig.mcpServers.deliberation;
-  mcpConfig.mcpServers.deliberation = {
-    command: "node",
-    args: [toForwardSlash(path.join(INSTALL_DIR, "index.js"))],
-  };
+    let mcpConfig = {};
+    if (fs.existsSync(MCP_CONFIG)) {
+      try {
+        mcpConfig = JSON.parse(fs.readFileSync(MCP_CONFIG, "utf-8"));
+      } catch {
+        mcpConfig = {};
+      }
+    }
+
+    if (!mcpConfig.mcpServers) mcpConfig.mcpServers = {};
+
+    const alreadyRegistered = !!mcpConfig.mcpServers.deliberation;
+    mcpConfig.mcpServers.deliberation = {
+      command: "node",
+      args: [serverEntryPoint],
+    };
 
-  fs.writeFileSync(MCP_CONFIG, JSON.stringify(mcpConfig, null, 2));
-  log(alreadyRegistered
-    ? "   → Existing registration updated"
-    : "   → Registered successfully");
+    fs.writeFileSync(MCP_CONFIG, JSON.stringify(mcpConfig, null, 2));
+    log(alreadyRegistered
+      ? "   → Fallback: existing registration updated in .mcp.json"
+      : "   → Fallback: registered in .mcp.json");
+    log("   ⚠️ Claude CLI not found. Run manually if server isn't detected:");
+    log(`     claude mcp add deliberation -s user -- node "${serverEntryPoint}"`);
+  }
 
   // Step 5: Register Gemini CLI MCP server
   log("🔧 Registering Gemini CLI MCP server...");
@@ -261,7 +293,7 @@ if (args.includes("--help") || args.includes("-h")) {
 Deliberation MCP Server Installer
 
 Usage:
-  npx --yes --package @dmsdc-ai/aigentry-deliberation deliberation-install
+  npx @dmsdc-ai/aigentry-deliberation install
   node install.js
 
 Options:
@@ -283,14 +315,24 @@ Skill path:   ${SKILL_DEST}
 } else if (args.includes("--uninstall") || args.includes("uninstall")) {
   console.log("\n🗑️ Deliberation MCP Server — Uninstalling\n");
 
-  // Remove from Claude MCP config
+  // Remove from Claude Code MCP registration
+  let claudeUnregistered = false;
+  if (commandExists("claude")) {
+    try {
+      execSync("claude mcp remove deliberation -s user", { stdio: "pipe" });
+      log("Claude Code MCP server unregistered via 'claude mcp remove'");
+      claudeUnregistered = true;
+    } catch { /* may not exist */ }
+  }
+
+  // Also clean up legacy .mcp.json fallback
   if (fs.existsSync(MCP_CONFIG)) {
     try {
       const mcpConfig = JSON.parse(fs.readFileSync(MCP_CONFIG, "utf-8"));
       if (mcpConfig.mcpServers?.deliberation) {
         delete mcpConfig.mcpServers.deliberation;
         fs.writeFileSync(MCP_CONFIG, JSON.stringify(mcpConfig, null, 2));
-        log("Claude Code MCP server unregistered");
+        if (!claudeUnregistered) log("Claude Code MCP server unregistered from .mcp.json");
       }
     } catch { /* ignore */ }
   }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@dmsdc-ai/aigentry-deliberation",
-  "version": "0.0.37",
+  "version": "0.0.38",
   "description": "MCP Deliberation Server — Multi-session AI deliberation with smart speaker ordering and persona roles",
   "type": "module",
   "license": "MIT",

package/skills/deliberation/SKILL.md

@@ -27,7 +27,6 @@ Claude/Codex를 포함해 MCP를 지원하는 임의 CLI들이 구조화된 토
 | `deliberation_speaker_candidates` | 참가 가능한 speaker 후보 목록 조회 | 불필요 |
 | `deliberation_list_active` | 진행 중인 모든 세션 목록 | 불필요 |
 | `deliberation_status` | 토론 상태 조회 | 선택적* |
-| `deliberation_inject_context` | 외부 실험 결과/추가 컨텍스트 주입 | 선택적 |
 | `deliberation_context` | 프로젝트 컨텍스트 로드 | 불필요 |
 | `deliberation_browser_llm_tabs` | 브라우저 LLM 탭 목록 (웹 기반 LLM 참여용) | 불필요 |
 | `deliberation_route_turn` | 현재 차례 speaker의 transport(CLI/browser_auto/manual)를 자동 라우팅 | 선택적* |
@@ -107,57 +106,7 @@ Claude/Codex를 포함해 MCP를 지원하는 임의 CLI들이 구조화된 토
 4. 각 세션을 `session_id`로 명시해 독립 진행
 5. 각각 `deliberation_synthesize`로 개별 종료
 
-### C. 실험 회고 / keep-discard review
-autoresearch 스타일 실험 루프를 검토할 때는 긴 컨텍스트를 `topic`에 섞지 말고, 시작 후 `deliberation_inject_context`로 compact bundle을 넣습니다.
-
-권장 규칙:
-- inject payload는 `1.5KB ~ 2KB` 목표
-- 최근 `3~5`개 실험만 포함
-- `key_changes`는 최대 `3`개 scalar before/after만 유지
-- 전체 `results.tsv` / 전체 `program.md`는 넣지 말고 artifact path만 남김
-
-예:
-```text
-1. deliberation_start(topic: "experiment retrospective / keep-discard review", ...)
-2. deliberation_inject_context(
-     session_id: "<session_id>",
-     speaker: "dustcraw",
-     context: JSON.stringify({
-       past_experiments: [{
-         experiment_id: "dg-20260310-001",
-         signal_kind: "INTEREST_DRIFT",
-         patch_summary: "Raised relevanceThreshold from 0.30 to 0.35",
-         patch_kind: "config",
-         key_changes: {
-           relevanceThreshold: { before: 0.30, after: 0.35 }
-         },
-         score: 0.08,
-         score_label: "promotion_rate_delta",
-         metric_name: "promotion_rate_delta",
-         metric_delta: 0.08,
-         verdict: "positive",
-         followup_action: "kept",
-         reasoning: "Threshold raise reduced noise; promotion quality improved 8%"
-       }],
-       experiment_count: 1,
-       success_rate: 1.0
-     })
-   )
-3. deliberation_route_turn(...) 반복
-4. deliberation_synthesize(..., structured: {
-     summary: "...",
-     decisions: ["..."],
-     actionable_tasks: [...],
-     experiment_outcome: {
-       verdict: "modify",
-       suggested_action: "iterate",
-       confidence: 0.78,
-       measurement_window_hours: 24
-     }
-   })
-```
-
-### D. 자동 진행 (스크립트)
+### C. 자동 진행 (스크립트)
 ```bash
 # 새 토론
 bash auto-deliberate.sh "저장소 전략"
@@ -169,7 +118,7 @@ bash auto-deliberate.sh "API 설계" 5
 bash auto-deliberate.sh --resume <session_id>
 ```
 
-### E. 모니터링
+### D. 모니터링
 ```bash
 # 모든 활성 세션 모니터링
 bash deliberation-monitor.sh
@@ -181,14 +130,14 @@ bash deliberation-monitor.sh <session_id>
 bash deliberation-monitor.sh --tmux
 ```
 
-### F. 브라우저 LLM 자동 연결 (CDP Auto-Activation)
+### E. 브라우저 LLM 자동 연결 (CDP Auto-Activation)
 - 브라우저 LLM speaker가 선택되면 CDP(Chrome DevTools Protocol)가 자동으로 활성화됩니다.
 - macOS에서는 Chrome이 실행되지 않은 경우 `--remote-debugging-port=9222`로 자동 실행을 시도합니다.
 - **Chrome이 이미 CDP 없이 실행 중인 경우**: Chrome을 완전히 종료한 후 다시 시도해야 합니다. (최초 1회만 필요)
 - CDP 연결 성공 시 모든 브라우저 speaker는 ⚡자동 모드로 동작합니다.
 - Windows/Linux에서는 사용자가 직접 Chrome을 `--remote-debugging-port=9222`로 실행해야 합니다.
 
-### G. Chrome 확장 프로그램 사이드패널 지원
+### F. Chrome 확장 프로그램 사이드패널 지원
 - **Chrome 확장 프로그램 사이드패널 (chrome-extension:// URL)은 지원됩니다.**
 - Claude, ChatGPT, Gemini 등의 Chrome 확장 사이드패널도 CDP를 통해 deliberation 참가자로 사용 가능합니다.
 - 사이드패널은 title 기반 매칭으로 감지됩니다 (extension ID가 아닌 탭 제목으로 식별).