diff --git a/.claude/agents/agentica-agent.md b/.claude/agents/agentica-agent.md
index b520a1da..c92fa758 100644
--- a/.claude/agents/agentica-agent.md
+++ b/.claude/agents/agentica-agent.md
@@ -14,7 +14,7 @@ You are a specialized agent for building Python agents using the Agentica SDK. Y
 Before starting, read the SDK skill for full API reference:
 
 ```bash
-cat $CLAUDE_PROJECT_DIR/.claude/skills/agentica-sdk/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/agentica-sdk/SKILL.md
 ```
 
 ## Step 2: Understand Your Task
diff --git a/.claude/agents/braintrust-analyst.md b/.claude/agents/braintrust-analyst.md
index 046859fe..e8ac8f6a 100644
--- a/.claude/agents/braintrust-analyst.md
+++ b/.claude/agents/braintrust-analyst.md
@@ -19,7 +19,7 @@ You are a specialized analysis agent. Your job is to run Braintrust analysis scr
 Read the braintrust-analyze skill:
 
 ```bash
-cat $CLAUDE_PROJECT_DIR/.claude/skills/braintrust-analyze/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/braintrust-analyze/SKILL.md
 ```
 
 ## Step 2: Execute Analysis
@@ -27,7 +27,7 @@ cat $CLAUDE_PROJECT_DIR/.claude/skills/braintrust-analyze/SKILL.md
 Run analysis IMMEDIATELY using Bash tool:
 
 ```bash
-cd $CLAUDE_PROJECT_DIR && uv run python -m runtime.harness scripts/braintrust_analyze.py --last-session
+cd $CLAUDE_OPC_DIR && uv run python -m runtime.harness scripts/braintrust_analyze.py --last-session
 ```
 
 Other analyses (run as needed):
diff --git a/.claude/agents/debug-agent.md b/.claude/agents/debug-agent.md
index a27a2e0c..92ffab8e 100644
--- a/.claude/agents/debug-agent.md
+++ b/.claude/agents/debug-agent.md
@@ -13,7 +13,7 @@ You are a specialized debugging agent. Your job is to investigate issues, trace
 Before starting, read the debug skill for methodology:
 
 ```bash
-cat $CLAUDE_PROJECT_DIR/.claude/skills/debug/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/debug/SKILL.md
 ```
 
 Follow the structure and guidelines from that skill.
diff --git a/.claude/agents/memory-extractor.md b/.claude/agents/memory-extractor.md
index c20edd25..aa59f791 100644
--- a/.claude/agents/memory-extractor.md
+++ b/.claude/agents/memory-extractor.md
@@ -31,7 +31,7 @@ You receive:
 
 ```bash
 # Use the extraction script with filtering
-(cd $CLAUDE_PROJECT_DIR/opc && uv run python scripts/core/extract_thinking_blocks.py \
+(cd $CLAUDE_OPC_DIR && uv run python scripts/core/extract_thinking_blocks.py \
   --jsonl "$JSONL_PATH" \
   --filter \
   --format json) > /tmp/perception-blocks.json
@@ -42,7 +42,7 @@ This extracts only thinking blocks containing perception signals (actually, real
 ### Step 2: Check Stats
 
 ```bash
-(cd $CLAUDE_PROJECT_DIR/opc && uv run python scripts/core/extract_thinking_blocks.py \
+(cd $CLAUDE_OPC_DIR && uv run python scripts/core/extract_thinking_blocks.py \
   --jsonl "$JSONL_PATH" \
   --stats)
 ```
@@ -80,7 +80,7 @@ For each extracted perception change, use the mapped type from Step 3:
 
 ```bash
 # Example for a CORRECTION → ERROR_FIX
-(cd $CLAUDE_PROJECT_DIR/opc && uv run python scripts/core/store_learning.py \
+(cd $CLAUDE_OPC_DIR && uv run python scripts/core/store_learning.py \
   --session-id "$SESSION_ID" \
   --type "ERROR_FIX" \
   --context "what this relates to" \
@@ -90,7 +90,7 @@ For each extracted perception change, use the mapped type from Step 3:
   --json)
 
 # Example for a REALIZATION/INSIGHT → CODEBASE_PATTERN
-(cd $CLAUDE_PROJECT_DIR/opc && uv run python scripts/core/store_learning.py \
+(cd $CLAUDE_OPC_DIR && uv run python scripts/core/store_learning.py \
   --session-id "$SESSION_ID" \
   --type "CODEBASE_PATTERN" \
   --context "what this relates to" \
@@ -100,7 +100,7 @@ For each extracted perception change, use the mapped type from Step 3:
   --json)
 
 # Example for a DEBUGGING_APPROACH → WORKING_SOLUTION
-(cd $CLAUDE_PROJECT_DIR/opc && uv run python scripts/core/store_learning.py \
+(cd $CLAUDE_OPC_DIR && uv run python scripts/core/store_learning.py \
   --session-id "$SESSION_ID" \
   --type "WORKING_SOLUTION" \
   --context "debugging methodology" \
diff --git a/.claude/agents/plan-agent.md b/.claude/agents/plan-agent.md
index f2446f92..3c29bf9d 100644
--- a/.claude/agents/plan-agent.md
+++ b/.claude/agents/plan-agent.md
@@ -13,7 +13,7 @@ You are a specialized planning agent. Your job is to create detailed implementat
 Before creating any plan, read the planning skill for methodology and format:
 
 ```bash
-cat $CLAUDE_PROJECT_DIR/.claude/skills/create_plan/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/create_plan/SKILL.md
 ```
 
 Follow the structure and guidelines from that skill.
diff --git a/.claude/agents/scribe.md b/.claude/agents/scribe.md
index 36150db2..53eef639 100644
--- a/.claude/agents/scribe.md
+++ b/.claude/agents/scribe.md
@@ -15,10 +15,10 @@ Before creating documentation, read the relevant skills:
 
 ```bash
 # For handoffs
-cat $CLAUDE_PROJECT_DIR/.claude/skills/create_handoff/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/create_handoff/SKILL.md
 
 # For ledger updates
-cat $CLAUDE_PROJECT_DIR/.claude/skills/continuity_ledger/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/continuity_ledger/SKILL.md
 ```
 
 Follow the structure and guidelines from those skills.
diff --git a/.claude/agents/session-analyst.md b/.claude/agents/session-analyst.md
index 81e0677b..68d965ce 100644
--- a/.claude/agents/session-analyst.md
+++ b/.claude/agents/session-analyst.md
@@ -13,7 +13,7 @@ You analyze Claude Code session data from Braintrust and provide insights.
 Read the skill file first:
 
 ```bash
-cat $CLAUDE_PROJECT_DIR/.claude/skills/braintrust-analyze/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/braintrust-analyze/SKILL.md
 ```
 
 ## Step 2: Run Analysis
diff --git a/.claude/agents/validate-agent.md b/.claude/agents/validate-agent.md
index 5b59eb32..fee65d9b 100644
--- a/.claude/agents/validate-agent.md
+++ b/.claude/agents/validate-agent.md
@@ -13,7 +13,7 @@ You are a specialized validation agent. Your job is to validate a technical plan
 Before validating, read the validation skill for methodology and format:
 
 ```bash
-cat $CLAUDE_PROJECT_DIR/.claude/skills/validate-agent/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/validate-agent/SKILL.md
 ```
 
 Follow the structure and guidelines from that skill.
diff --git a/.claude/hooks/dist/mcp-directory-guard.mjs b/.claude/hooks/dist/mcp-directory-guard.mjs
new file mode 100644
index 00000000..e50ce092
--- /dev/null
+++ b/.claude/hooks/dist/mcp-directory-guard.mjs
@@ -0,0 +1,86 @@
+// src/mcp-directory-guard.ts
+import { readFileSync } from "fs";
+
+// src/shared/opc-path.ts
+import { existsSync } from "fs";
+import { join } from "path";
+function getOpcDir() {
+  const envOpcDir = process.env.CLAUDE_OPC_DIR;
+  if (envOpcDir && existsSync(envOpcDir)) {
+    return envOpcDir;
+  }
+  const projectDir = process.env.CLAUDE_PROJECT_DIR || process.cwd();
+  const localOpc = join(projectDir, "opc");
+  if (existsSync(localOpc)) {
+    return localOpc;
+  }
+  const homeDir = process.env.HOME || process.env.USERPROFILE || "";
+  if (homeDir) {
+    const globalClaude = join(homeDir, ".claude");
+    const globalScripts = join(globalClaude, "scripts", "core");
+    if (existsSync(globalScripts)) {
+      return globalClaude;
+    }
+  }
+  return null;
+}
+
+// src/mcp-directory-guard.ts
+var SCRIPT_PATH_PATTERN = /\bscripts\/(mcp|core)\//;
+function buildCdPrefixPattern(opcDir) {
+  const escapedDir = opcDir ? opcDir.replace(/[.*+?^${}()|[\]\\]/g, "\\$&") : "";
+  const variants = [
+    "\\$CLAUDE_OPC_DIR",
+    "\\$\\{CLAUDE_OPC_DIR\\}"
+  ];
+  if (escapedDir) {
+    variants.push(escapedDir);
+  }
+  return new RegExp(`^\\s*cd\\s+(${variants.join("|")})\\s*&&`);
+}
+function main() {
+  let input;
+  try {
+    const stdinContent = readFileSync(0, "utf-8");
+    input = JSON.parse(stdinContent);
+  } catch {
+    console.log("{}");
+    return;
+  }
+  if (input.tool_name !== "Bash") {
+    console.log("{}");
+    return;
+  }
+  const command = input.tool_input?.command;
+  if (!command) {
+    console.log("{}");
+    return;
+  }
+  if (!SCRIPT_PATH_PATTERN.test(command)) {
+    console.log("{}");
+    return;
+  }
+  const opcDir = getOpcDir();
+  const cdPrefix = buildCdPrefixPattern(opcDir);
+  if (cdPrefix.test(command)) {
+    console.log("{}");
+    return;
+  }
+  const dirRef = opcDir || "$CLAUDE_OPC_DIR";
+  const corrected = `cd ${dirRef} && ${command.trimStart()}`;
+  const output = {
+    hookSpecificOutput: {
+      hookEventName: "PreToolUse",
+      permissionDecision: "deny",
+      permissionDecisionReason: `OPC directory guard: commands referencing scripts/(mcp|core)/ must run from the OPC directory so uv can find pyproject.toml.
+
+Blocked command:
+  ${command.trim()}
+
+Corrected command:
+  ${corrected}`
+    }
+  };
+  console.log(JSON.stringify(output));
+}
+main();
diff --git a/.claude/hooks/dist/session-start-continuity.mjs b/.claude/hooks/dist/session-start-continuity.mjs
index 671d6630..4e9ae9b6 100644
--- a/.claude/hooks/dist/session-start-continuity.mjs
+++ b/.claude/hooks/dist/session-start-continuity.mjs
@@ -1,7 +1,8 @@
 // src/session-start-continuity.ts
 import * as fs from "fs";
 import * as path from "path";
-import { execSync } from "child_process";
+import * as os from "os";
+import { execSync, spawn } from "child_process";
 function buildHandoffDirName(sessionName, sessionId) {
   const uuidShort = sessionId.replace(/-/g, "").slice(0, 8);
   return `${sessionName}-${uuidShort}`;
@@ -159,9 +160,42 @@ function getUnmarkedHandoffs() {
     return [];
   }
 }
+function ensureMemoryDaemon() {
+  const pidFile = path.join(os.homedir(), ".claude", "memory-daemon.pid");
+  if (fs.existsSync(pidFile)) {
+    try {
+      const pid = parseInt(fs.readFileSync(pidFile, "utf-8").trim(), 10);
+      process.kill(pid, 0);
+      return null;
+    } catch {
+      fs.unlinkSync(pidFile);
+    }
+  }
+  const possibleLocations = [
+    path.join(os.homedir(), ".claude", "opc", "scripts", "core", "memory_daemon.py"),
+    path.join(os.homedir(), ".claude", "scripts", "core", "memory_daemon.py")
+  ];
+  for (const daemonScript of possibleLocations) {
+    if (fs.existsSync(daemonScript)) {
+      try {
+        const child = spawn("uv", ["run", "python", daemonScript, "start"], {
+          detached: true,
+          stdio: "ignore",
+          cwd: path.dirname(path.dirname(path.dirname(daemonScript)))
+        });
+        child.unref();
+        return "Memory daemon: Started";
+      } catch (e) {
+        console.error(`Warning: Failed to start memory daemon: ${e}`);
+      }
+    }
+  }
+  return null;
+}
 async function main() {
   const input = JSON.parse(await readStdin());
   const projectDir = process.env.CLAUDE_PROJECT_DIR || process.cwd();
+  ensureMemoryDaemon();
   const sessionType = input.source || input.type;
   let message = "";
   let additionalContext = "";
diff --git a/.claude/hooks/dist/tldr-read-enforcer.mjs b/.claude/hooks/dist/tldr-read-enforcer.mjs
index 2aee9800..9898c839 100644
--- a/.claude/hooks/dist/tldr-read-enforcer.mjs
+++ b/.claude/hooks/dist/tldr-read-enforcer.mjs
@@ -1,6 +1,6 @@
 // src/tldr-read-enforcer.ts
 import { readFileSync as readFileSync2, existsSync as existsSync2, statSync } from "fs";
-import { basename, extname } from "path";
+import { basename, extname, isAbsolute, join as join2 } from "path";
 
 // src/daemon-client.ts
 import { existsSync, readFileSync, writeFileSync, unlinkSync } from "fs";
@@ -520,7 +520,8 @@ async function main() {
     console.log("{}");
     return;
   }
-  const filePath = input.tool_input.file_path || "";
+  const rawFilePath = input.tool_input.file_path || "";
+  const filePath = isAbsolute(rawFilePath) ? rawFilePath : join2(input.cwd, rawFilePath);
   if (!isCodeFile(filePath)) {
     console.log("{}");
     return;
diff --git a/.claude/hooks/src/mcp-directory-guard.ts b/.claude/hooks/src/mcp-directory-guard.ts
new file mode 100644
index 00000000..6709b191
--- /dev/null
+++ b/.claude/hooks/src/mcp-directory-guard.ts
@@ -0,0 +1,104 @@
+/**
+ * PreToolUse:Bash Hook - OPC Script Directory Guard
+ *
+ * Prevents running scripts from `scripts/(mcp|core)/` without first
+ * changing to $CLAUDE_OPC_DIR. When Claude runs these scripts from the
+ * wrong directory, `uv run` misses `opc/pyproject.toml` and its
+ * dependencies, causing ModuleNotFoundError.
+ *
+ * Detection: any Bash command referencing `scripts/(mcp|core)/` paths
+ * Allowed: commands prefixed with `cd $CLAUDE_OPC_DIR &&` (or resolved path)
+ * Denied: returns corrected command in the reason message
+ *
+ * Fixes: #148
+ */
+
+import { readFileSync } from 'fs';
+import { getOpcDir } from './shared/opc-path.js';
+import type { PreToolUseInput, PreToolUseHookOutput } from './shared/types.js';
+
+/**
+ * Pattern matching scripts/(mcp|core)/ references in Bash commands.
+ * Captures the path for use in the corrected command suggestion.
+ */
+const SCRIPT_PATH_PATTERN = /\bscripts\/(mcp|core)\//;
+
+/**
+ * Pattern matching a proper cd prefix to OPC dir.
+ * Accepts:
+ *   cd $CLAUDE_OPC_DIR &&
+ *   cd ${CLAUDE_OPC_DIR} &&
+ *   cd /resolved/opc/path &&
+ */
+function buildCdPrefixPattern(opcDir: string | null): RegExp {
+  const escapedDir = opcDir ? opcDir.replace(/[.*+?^${}()|[\]\\]/g, '\\$&') : '';
+  // Match: cd <opc-dir-variant> && (with flexible whitespace)
+  const variants = [
+    '\\$CLAUDE_OPC_DIR',
+    '\\$\\{CLAUDE_OPC_DIR\\}',
+  ];
+  if (escapedDir) {
+    variants.push(escapedDir);
+  }
+  return new RegExp(`^\\s*cd\\s+(${variants.join('|')})\\s*&&`);
+}
+
+function main(): void {
+  let input: PreToolUseInput;
+  try {
+    const stdinContent = readFileSync(0, 'utf-8');
+    input = JSON.parse(stdinContent) as PreToolUseInput;
+  } catch {
+    // Can't read input - allow through
+    console.log('{}');
+    return;
+  }
+
+  // Only process Bash tool
+  if (input.tool_name !== 'Bash') {
+    console.log('{}');
+    return;
+  }
+
+  const command = input.tool_input?.command as string;
+  if (!command) {
+    console.log('{}');
+    return;
+  }
+
+  // Check if command references OPC script paths
+  if (!SCRIPT_PATH_PATTERN.test(command)) {
+    // No script path reference - allow through
+    console.log('{}');
+    return;
+  }
+
+  const opcDir = getOpcDir();
+  const cdPrefix = buildCdPrefixPattern(opcDir);
+
+  // Check if command already has the correct cd prefix
+  if (cdPrefix.test(command)) {
+    console.log('{}');
+    return;
+  }
+
+  // Build corrected command suggestion
+  const dirRef = opcDir || '$CLAUDE_OPC_DIR';
+  const corrected = `cd ${dirRef} && ${command.trimStart()}`;
+
+  const output: PreToolUseHookOutput = {
+    hookSpecificOutput: {
+      hookEventName: 'PreToolUse',
+      permissionDecision: 'deny',
+      permissionDecisionReason:
+        `OPC directory guard: commands referencing scripts/(mcp|core)/ must ` +
+        `run from the OPC directory so uv can find pyproject.toml.\n\n` +
+        `Blocked command:\n  ${command.trim()}\n\n` +
+        `Corrected command:\n  ${corrected}`,
+    },
+  };
+
+  console.log(JSON.stringify(output));
+}
+
+main();
diff --git a/.claude/hooks/src/session-start-continuity.ts b/.claude/hooks/src/session-start-continuity.ts
index 8becbf59..e6a60a61 100644
--- a/.claude/hooks/src/session-start-continuity.ts
+++ b/.claude/hooks/src/session-start-continuity.ts
@@ -1,6 +1,7 @@
 import * as fs from 'fs';
 import * as path from 'path';
-import { execSync } from 'child_process';
+import * as os from 'os';
+import { execSync, spawn } from 'child_process';
 
 interface SessionStartInput {
   type?: 'startup' | 'resume' | 'clear' | 'compact';  // Legacy field
@@ -308,10 +309,53 @@ function getUnmarkedHandoffs(): UnmarkedHandoff[] {
   }
 }
 
+// ============================================
+// MEMORY DAEMON: Auto-start on session start
+// ============================================
+
+function ensureMemoryDaemon(): string | null {
+  const pidFile = path.join(os.homedir(), '.claude', 'memory-daemon.pid');
+
+  if (fs.existsSync(pidFile)) {
+    try {
+      const pid = parseInt(fs.readFileSync(pidFile, 'utf-8').trim(), 10);
+      process.kill(pid, 0);
+      return null;
+    } catch {
+      fs.unlinkSync(pidFile);
+    }
+  }
+
+  const possibleLocations = [
+    path.join(os.homedir(), '.claude', 'opc', 'scripts', 'core', 'memory_daemon.py'),
+    path.join(os.homedir(), '.claude', 'scripts', 'core', 'memory_daemon.py'),
+  ];
+
+  for (const daemonScript of possibleLocations) {
+    if (fs.existsSync(daemonScript)) {
+      try {
+        const child = spawn('uv', ['run', 'python', daemonScript, 'start'], {
+          detached: true,
+          stdio: 'ignore',
+          cwd: path.dirname(path.dirname(path.dirname(daemonScript))),
+        });
+        child.unref();
+        return 'Memory daemon: Started';
+      } catch (e) {
+        console.error(`Warning: Failed to start memory daemon: ${e}`);
+      }
+    }
+  }
+
+  return null;
+}
+
 async function main() {
   const input: SessionStartInput = JSON.parse(await readStdin());
   const projectDir = process.env.CLAUDE_PROJECT_DIR || process.cwd();
 
+  ensureMemoryDaemon();
+
   // Support both 'source' (per docs) and 'type' (legacy) fields
   const sessionType = input.source || input.type;
 
diff --git a/.claude/hooks/src/tldr-read-enforcer.ts b/.claude/hooks/src/tldr-read-enforcer.ts
index 3090da00..90a1cad5 100644
--- a/.claude/hooks/src/tldr-read-enforcer.ts
+++ b/.claude/hooks/src/tldr-read-enforcer.ts
@@ -10,7 +10,7 @@
  */
 
 import { readFileSync, existsSync, statSync } from 'fs';
-import { basename, extname } from 'path';
+import { basename, extname, isAbsolute, join } from 'path';
 import { queryDaemonSync, DaemonResponse, trackHookActivitySync } from './daemon-client';
 
 // Search context from smart-search-router
@@ -371,7 +371,8 @@ async function main() {
     return;
   }
 
-  const filePath = input.tool_input.file_path || '';
+  const rawFilePath = input.tool_input.file_path || '';
+  const filePath = isAbsolute(rawFilePath) ? rawFilePath : join(input.cwd, rawFilePath);
 
   // Allow non-code files
   if (!isCodeFile(filePath)) {
diff --git a/.claude/scripts/status.py b/.claude/scripts/status.py
index 04ee7f61..ce794dd9 100644
--- a/.claude/scripts/status.py
+++ b/.claude/scripts/status.py
@@ -1,8 +1,32 @@
 #!/usr/bin/env python3
 """Cross-platform status line for Claude Code.
 
-Shows: 145K 72% | main U:6 | Goal → Current focus
-Critical: ⚠ 160K 80% | main U:6 | Current focus
+Output Format:
+    CONTEXT: {tokens} {context}% | GIT: {branch} S:{staged} U:{unstaged} A:{added} | GOAL: {goal} → NOW: {now}
+
+Examples:
+    CONTEXT: 78K 39% | GIT: main A:4 | GOAL: Fix wizard → NOW: Update timeout
+    CONTEXT: 145K 72% | GIT: main S:2 U:1 | GOAL: Refactor OPC → NOW: Add validation
+    ⚠ CONTEXT: 160K 80% | GIT: dev U:6 | NOW: Critical bug fix
+
+Color Coding:
+    Green  (ctx < 60%): Normal operation
+    Yellow (ctx 60-79%): Warning, consider cleanup
+    Red    (ctx ≥ 80%): Critical, show only now focus
+
+Components:
+    tokens  - Total tokens used (input + cache_read + cache_creation + overhead)
+    context% - Percentage of context window consumed
+    branch  - Git branch name (max 12 chars)
+    S:N     - Staged files (git add done)
+    U:N     - Unstaged files (changed but not staged)
+    A:N     - Untracked files (new files not git-added)
+    goal    - Objective from handoff file (max 25 chars)
+    now     - Current focus from handoff (max 30 chars)
+
+Temporary Files:
+    ~/.tmp/claude-context-pct-{session_id}.txt  - Context % for other hooks
+    ~/.tmp/claude-session-stats-{session_id}.json - Full stats for /tldr-stats skill
 
 Replaces status.sh for Windows compatibility.
 """
@@ -89,11 +113,14 @@ def log_context_drop(session_id: str, prev_pct: int, curr_pct: int) -> None:
     Logs to ~/.claude/autocompact.log (local, not pushed to repo).
     """
     from datetime import datetime
+
     log_file = Path.home() / ".claude" / "autocompact.log"
     try:
         with open(log_file, "a") as f:
             timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
-            f.write(f"{timestamp} | session:{session_id} | {prev_pct}% → {curr_pct}% (drop: {prev_pct - curr_pct}%)\n")
+            f.write(
+                f"{timestamp} | session:{session_id} | {prev_pct}% → {curr_pct}% (drop: {prev_pct - curr_pct}%)\n"
+            )
     except OSError:
         pass
 
@@ -180,15 +207,27 @@ def get_git_info(cwd: Path) -> str:
         # Check if git repo
         result = subprocess.run(
             ["git", "-C", str(cwd), "rev-parse", "--git-dir"],
-            capture_output=True, text=True, timeout=5
+            capture_output=True,
+            text=True,
+            timeout=5,
         )
         if result.returncode != 0:
             return ""
 
         # Get branch name
         result = subprocess.run(
-            ["git", "-C", str(cwd), "--no-optional-locks", "rev-parse", "--abbrev-ref", "HEAD"],
-            capture_output=True, text=True, timeout=5
+            [
+                "git",
+                "-C",
+                str(cwd),
+                "--no-optional-locks",
+                "rev-parse",
+                "--abbrev-ref",
+                "HEAD",
+            ],
+            capture_output=True,
+            text=True,
+            timeout=5,
         )
         branch = result.stdout.strip() if result.returncode == 0 else ""
         if len(branch) > 12:
@@ -196,22 +235,46 @@ def get_git_info(cwd: Path) -> str:
 
         # Get staged count
         result = subprocess.run(
-            ["git", "-C", str(cwd), "--no-optional-locks", "diff", "--cached", "--name-only"],
-            capture_output=True, text=True, timeout=5
+            [
+                "git",
+                "-C",
+                str(cwd),
+                "--no-optional-locks",
+                "diff",
+                "--cached",
+                "--name-only",
+            ],
+            capture_output=True,
+            text=True,
+            timeout=5,
         )
         staged = len(result.stdout.strip().split("\n")) if result.stdout.strip() else 0
 
         # Get unstaged count
         result = subprocess.run(
             ["git", "-C", str(cwd), "--no-optional-locks", "diff", "--name-only"],
-            capture_output=True, text=True, timeout=5
+            capture_output=True,
+            text=True,
+            timeout=5,
+        )
+        unstaged = (
+            len(result.stdout.strip().split("\n")) if result.stdout.strip() else 0
         )
-        unstaged = len(result.stdout.strip().split("\n")) if result.stdout.strip() else 0
 
         # Get untracked count
         result = subprocess.run(
-            ["git", "-C", str(cwd), "--no-optional-locks", "ls-files", "--others", "--exclude-standard"],
-            capture_output=True, text=True, timeout=5
+            [
+                "git",
+                "-C",
+                str(cwd),
+                "--no-optional-locks",
+                "ls-files",
+                "--others",
+                "--exclude-standard",
+            ],
+            capture_output=True,
+            text=True,
+            timeout=5,
         )
         added = len(result.stdout.strip().split("\n")) if result.stdout.strip() else 0
 
@@ -239,8 +302,8 @@ def parse_filename_timestamp(path: Path) -> str:
 
     Returns '0000-00-00_00-00' if no timestamp found (sorts oldest).
     """
-    match = re.search(r'(\d{4}-\d{2}-\d{2}_\d{2}-\d{2})', path.name)
-    return match.group(1) if match else '0000-00-00_00-00'
+    match = re.search(r"(\d{4}-\d{2}-\d{2}_\d{2}-\d{2})", path.name)
+    return match.group(1) if match else "0000-00-00_00-00"
 
 
 def find_latest_handoff(project_dir: Path) -> Path | None:
@@ -267,7 +330,7 @@ def extract_yaml_field(content: str, field: str) -> str:
     pattern = rf"^{field}:\s*(.+?)$"
     match = re.search(pattern, content, re.MULTILINE)
     if match:
-        return match.group(1).strip().strip('"\'')
+        return match.group(1).strip().strip("\"'")
     return ""
 
 
@@ -300,14 +363,19 @@ def get_continuity_info(project_dir: Path) -> tuple[str, str]:
 
             # Fallback for now: Action Items or Next Steps section
             if not now_focus:
-                match = re.search(r"^## (?:Action Items|Next Steps)\s*\n(?:.*\n)*?^(\d+\.)\s*(.+?)$",
-                                  content, re.MULTILINE)
+                match = re.search(
+                    r"^## (?:Action Items|Next Steps)\s*\n(?:.*\n)*?^(\d+\.)\s*(.+?)$",
+                    content,
+                    re.MULTILINE,
+                )
                 if match:
                     now_focus = match.group(2).strip()
 
             # Try P0 section
             if not now_focus:
-                match = re.search(r"^### P0\s*\n(?:.*\n)*?^(\d+\.)\s*(.+?)$", content, re.MULTILINE)
+                match = re.search(
+                    r"^### P0\s*\n(?:.*\n)*?^(\d+\.)\s*(.+?)$", content, re.MULTILINE
+                )
                 if match:
                     now_focus = match.group(2).strip()
 
@@ -346,42 +414,43 @@ def get_continuity_info(project_dir: Path) -> tuple[str, str]:
     return goal, now_focus
 
 
-def build_output(context_pct: int, token_display: str, git_info: str,
-                 goal: str, now_focus: str) -> str:
+def build_output(
+    context_pct: int, token_display: str, git_info: str, goal: str, now_focus: str
+) -> str:
     """Build the final colored output string."""
     # Build continuity string
     if goal and now_focus:
-        continuity = f"{goal} → {now_focus}"
+        continuity = f"GOAL: {goal} → NOW: {now_focus}"
     elif now_focus:
-        continuity = now_focus
+        continuity = f"NOW: {now_focus}"
     elif goal:
-        continuity = goal
+        continuity = f"GOAL: {goal}"
     else:
         continuity = ""
 
     # Color based on context usage
     if context_pct >= 80:
         # CRITICAL - Red warning
-        ctx_display = f"\033[31m⚠ {token_display} {context_pct}%\033[0m"
+        ctx_display = f"\033[31m⚠ CONTEXT: {token_display} {context_pct}%\033[0m"
         parts = [ctx_display]
         if git_info:
-            parts.append(git_info)
+            parts.append(f"GIT: {git_info}")
         if now_focus:  # Only show now_focus when critical
-            parts.append(now_focus)
+            parts.append(continuity)
     elif context_pct >= 60:
         # WARNING - Yellow
-        ctx_display = f"\033[33m{token_display} {context_pct}%\033[0m"
+        ctx_display = f"\033[33mCONTEXT: {token_display} {context_pct}%\033[0m"
         parts = [ctx_display]
         if git_info:
-            parts.append(git_info)
+            parts.append(f"GIT: {git_info}")
         if continuity:
             parts.append(continuity)
     else:
         # NORMAL - Green
-        ctx_display = f"\033[32m{token_display} {context_pct}%\033[0m"
+        ctx_display = f"\033[32mCONTEXT: {token_display} {context_pct}%\033[0m"
         parts = [ctx_display]
         if git_info:
-            parts.append(git_info)
+            parts.append(f"GIT: {git_info}")
         if continuity:
             parts.append(continuity)
 
diff --git a/.claude/settings.json b/.claude/settings.json
index c575d685..ddb3536e 100644
--- a/.claude/settings.json
+++ b/.claude/settings.json
@@ -228,6 +228,11 @@
       {
         "matcher": "Bash",
         "hooks": [
+          {
+            "type": "command",
+            "command": "node $HOME/.claude/hooks/dist/mcp-directory-guard.mjs",
+            "timeout": 5
+          },
           {
             "type": "command",
             "command": "node $HOME/.claude/hooks/dist/import-error-detector.mjs",
diff --git a/.claude/skills/async-repl-protocol/SKILL.md b/.claude/skills/async-repl-protocol/SKILL.md
deleted file mode 100644
index 943d63f4..00000000
--- a/.claude/skills/async-repl-protocol/SKILL.md
+++ /dev/null
@@ -1,30 +0,0 @@
----
-name: async-repl-protocol
-description: Async REPL Protocol
-user-invocable: false
----
-
-# Async REPL Protocol
-
-When working with Agentica's async REPL harness for testing.
-
-## Rules
-
-### 1. Use `await` for Future-returning tools
-
-```python
-content = await view_file(path)  # NOT view_file(path)
-answer = await ask_memory("...")
-```
-
-### 2. Single code block per response
-
-Compute AND return in ONE block. Multiple blocks means only first executes.
-
-```python
-# GOOD: Single block
-content = await view_file(path)
-return any(c.isdigit() for c in content)
-
-# BAD: Split blocks (second block never runs)
-content = await view_file(path)
diff --git a/.claude/skills/commit/SKILL.md b/.claude/skills/commit/SKILL.md
index 1a3c8e35..25cc0081 100644
--- a/.claude/skills/commit/SKILL.md
+++ b/.claude/skills/commit/SKILL.md
@@ -32,7 +32,7 @@ You are tasked with creating git commits for the changes made during this sessio
    - Show the result with `git log --oneline -n [number]`
 
 5. **Generate reasoning (after each commit):**
-   - Run: `bash "$CLAUDE_PROJECT_DIR/.claude/scripts/generate-reasoning.sh" <commit-hash> "<commit-message>"`
+   - Run: `bash "$CLAUDE_CC_DIR/.claude/scripts/generate-reasoning.sh" <commit-hash> "<commit-message>"`
    - This captures what was tried during development (build failures, fixes)
    - The reasoning file helps future sessions understand past decisions
    - Stored in `.git/claude/commits/<hash>/reasoning.md`
diff --git a/.claude/skills/commit/SKILL.v6.md b/.claude/skills/commit/SKILL.v6.md
index 870a2dd2..735dbc32 100644
--- a/.claude/skills/commit/SKILL.v6.md
+++ b/.claude/skills/commit/SKILL.v6.md
@@ -71,7 +71,7 @@ eta |-> generate_reasoning(hash, message)
 ```bash
 git add <file1> <file2> ...
 git commit -m "message"
-bash "$CLAUDE_PROJECT_DIR/.claude/scripts/generate-reasoning.sh" <hash> "<message>"
+bash "$CLAUDE_CC_DIR/.claude/scripts/generate-reasoning.sh" <hash> "<message>"
 git log --oneline -n N
 ```
 
diff --git a/.claude/skills/continuity_ledger/SKILL.md b/.claude/skills/continuity-ledger/SKILL.md
similarity index 100%
rename from .claude/skills/continuity_ledger/SKILL.md
rename to .claude/skills/continuity-ledger/SKILL.md
diff --git a/.claude/skills/create_handoff/SKILL.md b/.claude/skills/create-handoff/SKILL.md
similarity index 100%
rename from .claude/skills/create_handoff/SKILL.md
rename to .claude/skills/create-handoff/SKILL.md
diff --git a/.claude/skills/create_handoff/SKILL.v6.md b/.claude/skills/create-handoff/SKILL.v6.md
similarity index 100%
rename from .claude/skills/create_handoff/SKILL.v6.md
rename to .claude/skills/create-handoff/SKILL.v6.md
diff --git a/.claude/skills/describe_pr/SKILL.md b/.claude/skills/describe-pr/SKILL.md
similarity index 97%
rename from .claude/skills/describe_pr/SKILL.md
rename to .claude/skills/describe-pr/SKILL.md
index fbf8c425..20074e87 100644
--- a/.claude/skills/describe_pr/SKILL.md
+++ b/.claude/skills/describe-pr/SKILL.md
@@ -34,7 +34,7 @@ You are tasked with generating a comprehensive pull request description followin
 
 4b. **Gather reasoning history (if available):**
    - Check if reasoning files exist: `ls .git/claude/commits/*/reasoning.md 2>/dev/null`
-   - If they exist, aggregate them: `bash "$CLAUDE_PROJECT_DIR/.claude/scripts/aggregate-reasoning.sh" main`
+   - If they exist, aggregate them: `bash "$CLAUDE_CC_DIR/.claude/scripts/aggregate-reasoning.sh" main`
    - This shows what approaches were tried before the final solution
    - Save the output for inclusion in the PR description
 
diff --git a/.claude/skills/discovery-interview/SKILL.md b/.claude/skills/discovery-interview/SKILL.md
index ce4ac8c0..517ce079 100644
--- a/.claude/skills/discovery-interview/SKILL.md
+++ b/.claude/skills/discovery-interview/SKILL.md
@@ -2,7 +2,6 @@
 name: discovery-interview
 description: Deep interview process to transform vague ideas into detailed specs. Works for technical and non-technical users.
 user-invocable: true
-model: claude-opus-4-5-20251101
 ---
 
 # Discovery Interview
diff --git a/.claude/skills/git-commits/SKILL.md b/.claude/skills/git-commits/SKILL.md
index cd390434..836aeedf 100644
--- a/.claude/skills/git-commits/SKILL.md
+++ b/.claude/skills/git-commits/SKILL.md
@@ -38,7 +38,7 @@ When you see these in user prompts, use the commit skill:
 
 The skill will prompt you to run:
 ```bash
-bash "$CLAUDE_PROJECT_DIR/.claude/scripts/generate-reasoning.sh" <hash> "<message>"
+bash "$CLAUDE_CC_DIR/.claude/scripts/generate-reasoning.sh" <hash> "<message>"
 ```
 
 Then push if requested:
diff --git a/.claude/skills/implement_plan/SKILL.md b/.claude/skills/implement-plan/SKILL.md
similarity index 99%
rename from .claude/skills/implement_plan/SKILL.md
rename to .claude/skills/implement-plan/SKILL.md
index db53ac4d..bf9b7ae8 100644
--- a/.claude/skills/implement_plan/SKILL.md
+++ b/.claude/skills/implement-plan/SKILL.md
@@ -1,5 +1,5 @@
 ---
-name: implement_plan
+name: implement-plan
 description: Implement technical plans from thoughts/shared/plans with verification
 user-invocable: false
 ---
diff --git a/.claude/skills/implement_plan/SKILL.md.backup b/.claude/skills/implement-plan/SKILL.md.backup
similarity index 100%
rename from .claude/skills/implement_plan/SKILL.md.backup
rename to .claude/skills/implement-plan/SKILL.md.backup
diff --git a/.claude/skills/implement_task/SKILL.md b/.claude/skills/implement-task/SKILL.md
similarity index 99%
rename from .claude/skills/implement_task/SKILL.md
rename to .claude/skills/implement-task/SKILL.md
index 248a4947..5f0d9483 100644
--- a/.claude/skills/implement_task/SKILL.md
+++ b/.claude/skills/implement-task/SKILL.md
@@ -1,5 +1,5 @@
 ---
-name: implement_task
+name: implement-task
 description: Implementation agent that executes a single task and creates handoff on completion
 user-invocable: false
 ---
diff --git a/.claude/skills/math-unified/SKILL.md b/.claude/skills/math-unified/SKILL.md
index 70d39795..7bf96c3c 100644
--- a/.claude/skills/math-unified/SKILL.md
+++ b/.claude/skills/math-unified/SKILL.md
@@ -31,7 +31,7 @@ For formal proofs, use `/prove` instead.
 
 ### SymPy (Symbolic Math)
 ```bash
-uv run python "$CLAUDE_PROJECT_DIR/.claude/scripts/cc_math/sympy_compute.py" <command> <args>
+uv run python "$CLAUDE_OPC_DIR/scripts/cc_math/sympy_compute.py" <command> <args>
 ```
 
 | Command | Description | Example |
@@ -83,7 +83,7 @@ uv run python "$CLAUDE_PROJECT_DIR/.claude/scripts/cc_math/sympy_compute.py" <co
 
 ### Z3 (Constraint Solving)
 ```bash
-uv run python "$CLAUDE_PROJECT_DIR/.claude/scripts/cc_math/z3_solve.py" <command> <args>
+uv run python "$CLAUDE_OPC_DIR/scripts/cc_math/z3_solve.py" <command> <args>
 ```
 
 | Command | Use Case |
@@ -96,7 +96,7 @@ uv run python "$CLAUDE_PROJECT_DIR/.claude/scripts/cc_math/z3_solve.py" <command
 
 ### Pint (Units)
 ```bash
-uv run python "$CLAUDE_PROJECT_DIR/.claude/scripts/cc_math/pint_compute.py" convert <value> <from_unit> <to_unit>
+uv run python "$CLAUDE_OPC_DIR/scripts/cc_math/pint_compute.py" convert <value> <from_unit> <to_unit>
 ```
 
 Example: `convert 5 miles kilometers`
@@ -105,7 +105,7 @@ Example: `convert 5 miles kilometers`
 
 ### Math Router (Auto-Route)
 ```bash
-uv run python "$CLAUDE_PROJECT_DIR/.claude/scripts/cc_math/math_router.py" route "<natural language request>"
+uv run python "$CLAUDE_OPC_DIR/scripts/cc_math/math_router.py" route "<natural language request>"
 ```
 
 Returns the exact command to run. Use when unsure which script.
@@ -154,28 +154,28 @@ I decide based on your request:
 ### Solve Equation
 ```
 User: Solve x² - 5x + 6 = 0
-Claude: uv run python "$CLAUDE_PROJECT_DIR/.claude/scripts/cc_math/sympy_compute.py" solve "x**2 - 5*x + 6" --var x
+Claude: uv run python "$CLAUDE_OPC_DIR/scripts/cc_math/sympy_compute.py" solve "x**2 - 5*x + 6" --var x
 Result: x = 2 or x = 3
 ```
 
 ### Compute Eigenvalues
 ```
 User: Find eigenvalues of [[2, 1], [1, 2]]
-Claude: uv run python "$CLAUDE_PROJECT_DIR/.claude/scripts/cc_math/sympy_compute.py" eigenvalues "[[2,1],[1,2]]"
+Claude: uv run python "$CLAUDE_OPC_DIR/scripts/cc_math/sympy_compute.py" eigenvalues "[[2,1],[1,2]]"
 Result: {1: 1, 3: 1}  (eigenvalue 1 with multiplicity 1, eigenvalue 3 with multiplicity 1)
 ```
 
 ### Prove Inequality
 ```
 User: Is x² + y² ≥ 2xy always true?
-Claude: uv run python "$CLAUDE_PROJECT_DIR/.claude/scripts/cc_math/z3_solve.py" prove "x**2 + y**2 >= 2*x*y"
+Claude: uv run python "$CLAUDE_OPC_DIR/scripts/cc_math/z3_solve.py" prove "x**2 + y**2 >= 2*x*y"
 Result: PROVED (equivalent to (x-y)² ≥ 0)
 ```
 
 ### Convert Units
 ```
 User: How many kilometers in 26.2 miles?
-Claude: uv run python "$CLAUDE_PROJECT_DIR/.claude/scripts/cc_math/pint_compute.py" convert 26.2 miles kilometers
+Claude: uv run python "$CLAUDE_OPC_DIR/scripts/cc_math/pint_compute.py" convert 26.2 miles kilometers
 Result: 42.16 km
 ```
 
diff --git a/.claude/skills/recall-reasoning/SKILL.md b/.claude/skills/recall-reasoning/SKILL.md
index 6110a74a..02c3dce5 100644
--- a/.claude/skills/recall-reasoning/SKILL.md
+++ b/.claude/skills/recall-reasoning/SKILL.md
@@ -32,7 +32,7 @@ This searches handoffs with post-mortems (what worked, what failed, key decision
 ### Secondary: Reasoning Files (build attempts)
 
 ```bash
-bash "$CLAUDE_PROJECT_DIR/.claude/scripts/search-reasoning.sh" "<query>"
+bash "$CLAUDE_CC_DIR/.claude/scripts/search-reasoning.sh" "<query>"
 ```
 
 This searches `.git/claude/commits/*/reasoning.md` for build failures and fixes.
@@ -50,7 +50,7 @@ uv run python scripts/core/artifact_query.py "implement agent" --outcome SUCCEED
 uv run python scripts/core/artifact_query.py "hook implementation" --outcome FAILED
 
 # Search build/test reasoning
-bash "$CLAUDE_PROJECT_DIR/.claude/scripts/search-reasoning.sh" "TypeError"
+bash "$CLAUDE_CC_DIR/.claude/scripts/search-reasoning.sh" "TypeError"
 ```
 
 ## What Gets Searched
diff --git a/.claude/skills/research/SKILL.md b/.claude/skills/research/SKILL.md
index 5ffec538..ab247ae7 100644
--- a/.claude/skills/research/SKILL.md
+++ b/.claude/skills/research/SKILL.md
@@ -1,7 +1,6 @@
 ---
 name: research
 description: Document codebase as-is with thoughts directory for historical context
-model: claude-opus-4-5-20251101
 user-invocable: false
 ---
 
diff --git a/.claude/skills/resume_handoff/SKILL.md b/.claude/skills/resume-handoff/SKILL.md
similarity index 100%
rename from .claude/skills/resume_handoff/SKILL.md
rename to .claude/skills/resume-handoff/SKILL.md
diff --git a/.claude/skills/router-first-architecture/SKILL.md b/.claude/skills/router-first-architecture/SKILL.md
deleted file mode 100644
index f4947d7b..00000000
--- a/.claude/skills/router-first-architecture/SKILL.md
+++ /dev/null
@@ -1,53 +0,0 @@
----
-name: router-first-architecture
-description: Router-First Architecture
-user-invocable: false
----
-
-# Router-First Architecture
-
-Route through domain routers before using individual tools. Routers abstract tool selection.
-
-## Pattern
-
-Domain routers (like `math-router`) provide deterministic mapping from user intent to exact CLI commands. Always use the router first; only bypass for edge cases.
-
-## DO
-
-- Call `math-router route "<intent>"` before any math operation
-- Let domain skills co-activate with their router (via `coActivate` in skill-rules.json)
-- Trust the router's confidence score; only fall back if `command: null`
-- Keep trigger keywords/patterns in skill-rules.json broader than routing patterns
-
-## DON'T
-
-- Call individual scripts directly when a router exists
-- Duplicate routing logic in individual skills
-- Let domain skills bypass their router
-
-## Co-Activation Pattern
-
-Domain skills should co-activate with their router:
-
-```json
-{
-  "math/abstract-algebra/groups": {
-    "coActivate": ["math-router"],
-    "coActivateMode": "always"
-  }
-}
-```
-
-This ensures the router is always available when domain knowledge is activated.
-
-## Two-Layer Architecture
-
-1. **Skill-rules trigger layer**: Nudges Claude to use the router (keywords, intent patterns)
-2. **Router routing layer**: Deterministic mapping to scripts via regex patterns
-
-Keep the trigger layer broader than routing - the router should handle "not found" gracefully.
-
-## Source Sessions
-
-- 2bbc8d6e: "Trigger layer was narrower than routing layer" - expanded triggers
-- This session: Wired 8 domain math skills to co-activate with math-router
diff --git a/.claude/skills/search-hierarchy/SKILL.md b/.claude/skills/search-hierarchy/SKILL.md
deleted file mode 100644
index 9fd1ee09..00000000
--- a/.claude/skills/search-hierarchy/SKILL.md
+++ /dev/null
@@ -1,85 +0,0 @@
----
-name: search-hierarchy
-description: Search Tool Hierarchy
-user-invocable: false
----
-
-# Search Tool Hierarchy
-
-Use the most token-efficient search tool for each query type.
-
-## Decision Tree
-
-```
-Query Type?
-├── STRUCTURAL (code patterns)
-│   → AST-grep (~50 tokens output)
-│   Examples: "def foo", "class Bar", "import X", "@decorator"
-│
-├── SEMANTIC (conceptual questions)
-│   → LEANN (~100 tokens if path-only)
-│   Examples: "how does auth work", "find error handling patterns"
-│
-├── LITERAL (exact identifiers)
-│   → Grep (variable output)
-│   Examples: "TemporalMemory", "check_evocation", regex patterns
-│
-└── FULL CONTEXT (need complete understanding)
-    → Read (1500+ tokens)
-    Last resort after finding the right file
-```
-
-## Token Efficiency Comparison
-
-| Tool | Output Size | Best For |
-|------|-------------|----------|
-| AST-grep | ~50 tokens | Function/class definitions, imports, decorators |
-| LEANN | ~100 tokens | Conceptual questions, architecture, patterns |
-| Grep | ~200-2000 | Exact identifiers, regex, file paths |
-| Read | ~1500+ | Full understanding after finding the file |
-
-## Hook Enforcement
-
-The `grep-to-leann.sh` hook automatically:
-1. Detects query type (structural/semantic/literal)
-2. Blocks and suggests AST-grep for structural queries
-3. Blocks and suggests LEANN for semantic queries
-4. Allows literal patterns through to Grep
-
-## DO
-
-- Start with AST-grep for code structure questions
-- Use LEANN for "how does X work" questions
-- Use Grep only for exact identifier matches
-- Read files only after finding them via search
-
-## DON'T
-
-- Use Grep for conceptual questions (returns nothing)
-- Read files before knowing which ones are relevant
-- Use Read when AST-grep would give file:line
-- Ignore hook suggestions
-
-## Examples
-
-```bash
-# STRUCTURAL → AST-grep
-ast-grep --pattern "async def $FUNC($$$):" --lang python
-
-# SEMANTIC → LEANN
-leann search opc-dev "how does authentication work" --top-k 3
-
-# LITERAL → Grep
-Grep pattern="check_evocation" path=opc/scripts
-
-# FULL CONTEXT → Read (after finding file)
-Read file_path=opc/scripts/z3_erotetic.py
-```
-
-## Optimal Flow
-
-```
-1. AST-grep: "Find async functions" → 3 file:line matches
-2. Read: Top match only → Full understanding
-3. Skip: 4 irrelevant files → 6000 tokens saved
-```
diff --git a/.claude/skills/search-tools/SKILL.md b/.claude/skills/search-tools/SKILL.md
deleted file mode 100644
index f9379606..00000000
--- a/.claude/skills/search-tools/SKILL.md
+++ /dev/null
@@ -1,72 +0,0 @@
----
-name: search-tools
-description: Search Tool Hierarchy
-user-invocable: false
----
-
-# Search Tool Hierarchy
-
-When searching code, use this decision tree:
-
-## Decision Tree
-
-```
-Need CONCEPTUAL/SEMANTIC search?
-  (how does X work, find patterns, understand architecture)
-  → Use LEANN (/leann-search) - embedding-based semantic search
-  → PreToolUse hook auto-redirects semantic Grep queries
-
-Need to understand code STRUCTURE?
-  (find function calls, class usages, refactor patterns)
-  → Use AST-grep (/ast-grep-find)
-
-Need to find TEXT in code?
-  → Use Morph (/morph-search) - 20x faster
-  → If no Morph API key: fall back to Grep tool
-
-Simple one-off search?
-  → Use built-in Grep tool directly
-```
-
-## Tool Comparison
-
-| Tool | Best For | Requires |
-|------|----------|----------|
-| **LEANN** | Semantic search: "how does caching work", "error handling patterns", conceptual queries | Index built |
-| **AST-grep** | Structural patterns: "find all calls to `foo()`", refactoring, find usages by type | MCP server |
-| **Morph** | Fast text search: "find files mentioning error", grep across codebase | API key |
-| **Grep** | Literal patterns, class/function names, regex | Nothing (built-in) |
-
-## Examples
-
-**LEANN** (semantic/conceptual):
-- "how does authentication work"
-- "find error handling patterns"
-- "where is rate limiting implemented"
-
-**AST-grep** (structural):
-- "Find all functions that return a Promise"
-- "Find all React components using useState"
-- "Refactor all imports of X to Y"
-
-**Morph** (text search):
-- "Find all files mentioning 'authentication'"
-- "Search for TODO comments"
-
-**Grep** (literal):
-- `class ProviderAdapter`
-- `def __init__`
-- Regex patterns
-
-## LEANN Commands
-
-```bash
-# Search with semantic query
-leann search opc-dev "how does blackboard communication work" --top-k 5
-
-# List available indexes
-leann list
-
-# Rebuild index (when code changes)
-leann build opc-dev --docs dir1 dir2 --no-recompute --no-compact --force
-```
diff --git a/.claude/skills/skill-developer/SKILL.md b/.claude/skills/skill-developer/SKILL.md
index de41038f..5678ab5d 100644
--- a/.claude/skills/skill-developer/SKILL.md
+++ b/.claude/skills/skill-developer/SKILL.md
@@ -56,14 +56,14 @@ To create a new MCP chain script and wrap it as a skill:
 Copy the multi-tool-pipeline template:
 
 ```bash
-cp $CLAUDE_PROJECT_DIR/scripts/multi_tool_pipeline.py $CLAUDE_PROJECT_DIR/scripts/my_pipeline.py
+cp $CLAUDE_OPC_DIR/scripts/multi_tool_pipeline.py $CLAUDE_PROJECT_DIR/scripts/my_pipeline.py
 ```
 
 Reference the template pattern:
 
 ```bash
-cat $CLAUDE_PROJECT_DIR/.claude/skills/multi-tool-pipeline/SKILL.md
-cat $CLAUDE_PROJECT_DIR/scripts/multi_tool_pipeline.py
+cat $CLAUDE_CC_DIR/.claude/skills/multi-tool-pipeline/SKILL.md
+cat $CLAUDE_OPC_DIR/scripts/multi_tool_pipeline.py
 ```
 
 ### Step 2: Customize the Script
@@ -140,8 +140,8 @@ Add to `.claude/skills/skill-rules.json`:
 For full details, read:
 
 ```bash
-cat $CLAUDE_PROJECT_DIR/.claude/rules/skill-development.md
-cat $CLAUDE_PROJECT_DIR/.claude/rules/mcp-scripts.md
+cat $CLAUDE_CC_DIR/.claude/rules/skill-development.md
+cat $CLAUDE_CC_DIR/.claude/rules/mcp-scripts.md
 ```
 
 ## Quick Checklist
@@ -157,7 +157,7 @@ cat $CLAUDE_PROJECT_DIR/.claude/rules/mcp-scripts.md
 Look at existing skills for patterns:
 
 ```bash
-ls $CLAUDE_PROJECT_DIR/.claude/skills/
-cat $CLAUDE_PROJECT_DIR/.claude/skills/commit/SKILL.md
-cat $CLAUDE_PROJECT_DIR/.claude/skills/firecrawl-scrape/SKILL.md
+ls $CLAUDE_CC_DIR/.claude/skills/
+cat $CLAUDE_CC_DIR/.claude/skills/commit/SKILL.md
+cat $CLAUDE_CC_DIR/.claude/skills/firecrawl-scrape/SKILL.md
 ```
diff --git a/.claude/skills/system_overview/SKILL.md b/.claude/skills/system-overview/SKILL.md
similarity index 100%
rename from .claude/skills/system_overview/SKILL.md
rename to .claude/skills/system-overview/SKILL.md
diff --git a/.claude/skills/tldr-stats/SKILL.md b/.claude/skills/tldr-stats/SKILL.md
index 6c16e4fd..cb2ed757 100644
--- a/.claude/skills/tldr-stats/SKILL.md
+++ b/.claude/skills/tldr-stats/SKILL.md
@@ -20,7 +20,7 @@ Show a beautiful dashboard with token usage, actual API costs, TLDR savings, and
 
 1. Run the stats script:
 ```bash
-python3 $CLAUDE_PROJECT_DIR/.claude/scripts/tldr_stats.py
+python3 $CLAUDE_CC_DIR/.claude/scripts/tldr_stats.py
 ```
 
 2. **Copy the full output into your response** so the user sees the dashboard directly in the chat. Do not just run the command silently - the user wants to see the stats.
diff --git a/opc/scripts/braintrust_analyze.py b/opc/scripts/braintrust_analyze.py
index 394f8c1d..f2cf0641 100644
--- a/opc/scripts/braintrust_analyze.py
+++ b/opc/scripts/braintrust_analyze.py
@@ -323,7 +323,7 @@ def list_sessions(project_id: str, api_key: str, limit: int = 5):
             MIN(created) as started,
             MAX(created) as ended,
             COUNT(*) as span_count,
-            COUNT(*) FILTER (WHERE span_attributes['type'] = 'tool') as tool_count
+            SUM(CASE WHEN span_attributes['type'] = 'tool' THEN 1 ELSE 0 END) as tool_count
         FROM logs
         GROUP BY root_span_id
         ORDER BY started DESC
diff --git a/opc/scripts/cc_math/math_base.py b/opc/scripts/cc_math/math_base.py
index 4a2f7526..0d2ca6b3 100644
--- a/opc/scripts/cc_math/math_base.py
+++ b/opc/scripts/cc_math/math_base.py
@@ -24,7 +24,7 @@
 from collections.abc import Callable
 from dataclasses import dataclass, field
 from functools import wraps
-from typing import Any, Dict, List, Optional, TypeVar, Union
+from typing import Any, TypeVar
 
 # Type variables for generic decorators
 T = TypeVar("T")
@@ -45,7 +45,7 @@ class MathCommand:
     category: str
     description: str
     latex_template: str | None = None
-    args: list[Dict[str, Any]] = field(default_factory=list)
+    args: list[dict[str, Any]] = field(default_factory=list)
 
 
 # Global registry per script - use module-level dict
@@ -57,7 +57,7 @@ def math_command(
     category: str,
     description: str = "",
     latex_template: str | None = None,
-    args: List[Dict[str, Any]] | None = None,
+    args: list[dict[str, Any]] | None = None,
 ) -> Callable[[F], F]:
     """Decorator to register a math command.
 
@@ -130,7 +130,7 @@ def clear_registry() -> None:
 # =============================================================================
 
 
-def format_output(result: dict[str, Any], latex_template: str | None = None) -> Dict[str, Any]:
+def format_output(result: dict[str, Any], latex_template: str | None = None) -> dict[str, Any]:
     """Format computation result as standardized JSON.
 
     Output structure:
@@ -719,7 +719,7 @@ def register_commands(
 
 def run_command(
     args: argparse.Namespace, registry: dict[str, MathCommand] | None = None
-) -> Dict[str, Any]:
+) -> dict[str, Any]:
     """Run command based on parsed arguments.
 
     Args:
@@ -764,15 +764,14 @@ def safe_compute(func: Callable, *args, timeout: int = 30, **kwargs) -> dict[str
     Returns:
         Result dict or error dict
     """
+    import multiprocessing
     from concurrent.futures import ProcessPoolExecutor
     from concurrent.futures import TimeoutError as FuturesTimeout
 
-    def _wrapper():
-        return func(*args, **kwargs)
-
+    ctx = multiprocessing.get_context("spawn")
     try:
-        with ProcessPoolExecutor(max_workers=1) as executor:
-            future = executor.submit(_wrapper)
+        with ProcessPoolExecutor(max_workers=1, mp_context=ctx) as executor:
+            future = executor.submit(func, *args, **kwargs)
             try:
                 return future.result(timeout=timeout)
             except FuturesTimeout:
diff --git a/opc/scripts/cc_math/sympy_compute.py b/opc/scripts/cc_math/sympy_compute.py
index 8e222e66..3fc51b33 100644
--- a/opc/scripts/cc_math/sympy_compute.py
+++ b/opc/scripts/cc_math/sympy_compute.py
@@ -1435,8 +1435,11 @@ def safe_solve(equation: str, var: str = "x", domain: str = "complex", timeout:
     Returns:
         dict with solutions on success, or error dict on failure/timeout
     """
+    import multiprocessing
+
+    ctx = multiprocessing.get_context("spawn")
     try:
-        with ProcessPoolExecutor(max_workers=1) as executor:
+        with ProcessPoolExecutor(max_workers=1, mp_context=ctx) as executor:
             future = executor.submit(_solve_internal, equation, var, domain)
             try:
                 return future.result(timeout=timeout)
diff --git a/opc/scripts/setup/claude_integration.py b/opc/scripts/setup/claude_integration.py
index 18deaf36..158d10c0 100644
--- a/opc/scripts/setup/claude_integration.py
+++ b/opc/scripts/setup/claude_integration.py
@@ -382,12 +382,12 @@ def generate_migration_guidance(
 
 # Root scripts used by skills/hooks - shared between copy and symlink install
 ROOT_SCRIPTS = [
-    "ast_grep_find.py",          # /ast-grep-find skill
-    "braintrust_analyze.py",     # session-end-cleanup hook
-    "qlty_check.py",             # /qlty-check skill
+    "ast_grep_find.py",  # /ast-grep-find skill
+    "braintrust_analyze.py",  # session-end-cleanup hook
+    "qlty_check.py",  # /qlty-check skill
     "research_implement_pipeline.py",  # /mcp-chaining skill
-    "test_research_pipeline.py", # /mcp-chaining skill
-    "multi_tool_pipeline.py",    # /skill-developer example
+    "test_research_pipeline.py",  # /mcp-chaining skill
+    "multi_tool_pipeline.py",  # /skill-developer example
     "recall_temporal_facts.py",  # /system_overview skill
 ]
 
@@ -431,6 +431,36 @@ def _copy_scripts(opc_source: Path, target_dir: Path) -> int:
     return count
 
 
+def _copy_dotfile_scripts(opc_source: Path, target_dir: Path) -> int:
+    """Copy root-level scripts from .claude/scripts/ to target.
+
+    These are scripts like status.py and tldr_stats.py that live in
+    .claude/scripts/ at the repo root, not in opc/scripts/.
+
+    Args:
+        opc_source: Source OPC .claude directory
+        target_dir: Target .claude directory
+
+    Returns:
+        Count of scripts copied
+    """
+    count = 0
+    source_scripts = opc_source.parent / ".claude" / "scripts"
+    target_scripts = target_dir / "scripts"
+    target_scripts.mkdir(parents=True, exist_ok=True)
+
+    if not source_scripts.exists():
+        return 0
+
+    for pattern in ["*.py", "*.sh"]:
+        for src in source_scripts.glob(pattern):
+            dst = target_scripts / src.name
+            shutil.copy2(src, dst)
+            count += 1
+
+    return count
+
+
 def install_opc_integration(
     target_dir: Path,
     opc_source: Path,
@@ -539,6 +569,9 @@ def install_opc_integration(
         # Copy scripts (core, math, tldr directories + root scripts)
         result["installed_scripts"] = _copy_scripts(opc_source, target_dir)
 
+        # Copy root-level scripts from .claude/scripts/ (e.g. status.py)
+        result["installed_scripts"] += _copy_dotfile_scripts(opc_source, target_dir)
+
         # Merge user items if requested
         if merge_user_items and existing and conflicts:
             # Merge non-conflicting hooks
@@ -670,6 +703,9 @@ def install_opc_integration_symlink(
         # Copy scripts (core, math, tldr directories + root scripts)
         _copy_scripts(opc_source, target_dir)
 
+        # Copy root-level scripts from .claude/scripts/ (e.g. status.py)
+        _copy_dotfile_scripts(opc_source, target_dir)
+
         result["success"] = True
 
     except Exception as e:
@@ -754,9 +790,7 @@ def strip_tldr_hooks_from_settings(settings_path: Path) -> bool:
                 if "startup" in matcher or "resume" in matcher:
                     hooks = hook_group.get("hooks", [])
                     new_hooks = [
-                        h
-                        for h in hooks
-                        if "session-start-tldr-cache" not in h.get("command", "")
+                        h for h in hooks if "session-start-tldr-cache" not in h.get("command", "")
                     ]
                     if len(new_hooks) != len(hooks):
                         modified = True
@@ -782,7 +816,6 @@ def strip_tldr_hooks_from_settings(settings_path: Path) -> bool:
         return False
 
 
-
 def get_platform_info() -> dict[str, str]:
     """Get current platform information.
 
@@ -812,15 +845,15 @@ def find_latest_backup(claude_dir: Path) -> Path | None:
 
 # Files to preserve during uninstall (user data accumulated since install)
 PRESERVE_FILES = [
-    "history.jsonl",      # Command history
-    "mcp_config.json",    # MCP server configs
-    ".env",               # API keys and settings
-    "projects.json",      # Project configs
+    "history.jsonl",  # Command history
+    "mcp_config.json",  # MCP server configs
+    ".env",  # API keys and settings
+    "projects.json",  # Project configs
 ]
 
 PRESERVE_DIRS = [
-    "file-history",       # File edit history
-    "projects",           # Project-specific data
+    "file-history",  # File edit history
+    "projects",  # Project-specific data
 ]
 
 
@@ -931,6 +964,32 @@ def uninstall_opc_integration(
     if result["preserved"]:
         msg_parts.append(f"  Preserved user data: {', '.join(result['preserved'])}")
 
+    # Clean up CLAUDE_OPC_DIR from shell configs
+    import os as _os
+
+    shell = _os.environ.get("SHELL", "")
+    configs_to_clean = []
+    if "zsh" in shell:
+        configs_to_clean.append(_os.path.expanduser("~/.zshrc"))
+    elif "bash" in shell:
+        configs_to_clean.append(_os.path.expanduser("~/.bashrc"))
+    elif "fish" in shell:
+        fish_config = _os.path.expanduser("~/.config/fish/config.fish")
+        if _os.path.exists(fish_config):
+            configs_to_clean.append(fish_config)
+
+    for config_path in configs_to_clean:
+        try:
+            with open(config_path, "r") as f:
+                lines = f.readlines()
+            new_lines = [l for l in lines if "CLAUDE_OPC_DIR" not in l]
+            if len(new_lines) < len(lines):
+                with open(config_path, "w") as f:
+                    f.writelines(new_lines)
+                msg_parts.append(f"  Removed CLAUDE_OPC_DIR from {_os.path.basename(config_path)}")
+        except Exception:
+            pass
+
     result["message"] = "\n".join(msg_parts)
     result["success"] = True
     return result
diff --git a/opc/scripts/setup/embedded_postgres.py b/opc/scripts/setup/embedded_postgres.py
index 4cc9f6b5..bedeb38b 100644
--- a/opc/scripts/setup/embedded_postgres.py
+++ b/opc/scripts/setup/embedded_postgres.py
@@ -17,12 +17,14 @@
 from urllib.parse import quote_plus
 
 
-def start_embedded_postgres(pgdata: Path) -> dict[str, Any]:
+def start_embedded_postgres(pgdata: Path, venv_path: Path | None = None) -> dict[str, Any]:
     """Start embedded postgres server using pgserver.
 
     Args:
         pgdata: Directory to store postgres data files.
                 Will be created if it doesn't exist.
+        venv_path: Optional path to venv with pgserver installed.
+                   If not provided, tries to import pgserver directly.
 
     Returns:
         dict with keys:
@@ -31,26 +33,52 @@ def start_embedded_postgres(pgdata: Path) -> dict[str, Any]:
             - error: str (if failed)
             - server: PostgresServer instance (for cleanup)
     """
-    try:
-        import pgserver
-    except ImportError:
-        return {
-            "success": False,
-            "error": "pgserver not installed. Install with: pip install pgserver",
-        }
+    import sys
 
-    try:
-        # Ensure pgdata directory exists
-        pgdata.mkdir(parents=True, exist_ok=True)
+    # Determine Python executable to use
+    if venv_path is not None:
+        if sys.platform == "win32":
+            python_exe = venv_path / "Scripts" / "python.exe"
+        else:
+            python_exe = venv_path / "bin" / "python"
+    else:
+        python_exe = sys.executable
+
+    # Build script to start pgserver and get URI
+    start_script = f"""
+import sys
+sys.path.insert(0, '{venv_path}/lib/python3.12/site-packages')
+from pgserver import get_server
+server = get_server('{pgdata}')
+print(server.get_uri())
+"""
 
-        # Start server (pgserver handles init if needed)
-        server = pgserver.get_server(str(pgdata))
-        uri = server.get_uri()
+    import subprocess
 
+    try:
+        proc = subprocess.Popen(
+            [str(python_exe), "-c", start_script],
+            stdout=subprocess.PIPE,
+            stderr=subprocess.PIPE,
+        )
+        stdout, stderr = proc.communicate(timeout=30)
+
+        if proc.returncode == 0:
+            uri = stdout.decode().strip()
+            return {
+                "success": True,
+                "uri": uri,
+            }
+        else:
+            error_msg = stderr.decode().strip() if stderr else "Unknown error"
+            return {
+                "success": False,
+                "error": f"Failed to start pgserver: {error_msg[:200]}",
+            }
+    except subprocess.TimeoutExpired:
         return {
-            "success": True,
-            "uri": uri,
-            "server": server,
+            "success": False,
+            "error": "pgserver start timed out",
         }
     except Exception as e:
         return {
@@ -260,7 +288,9 @@ async def setup_embedded_environment() -> dict[str, Any]:
         if venv_path.exists() and python_exe.exists():
             # Verify pgserver is installed
             proc = await asyncio.create_subprocess_exec(
-                str(python_exe), "-c", "import pgserver",
+                str(python_exe),
+                "-c",
+                "import pgserver",
                 stdout=asyncio.subprocess.PIPE,
                 stderr=asyncio.subprocess.PIPE,
             )
@@ -270,7 +300,11 @@ async def setup_embedded_environment() -> dict[str, Any]:
 
         # Create venv with Python 3.12
         proc = await asyncio.create_subprocess_exec(
-            "uv", "venv", str(venv_path), "--python", "3.12",
+            "uv",
+            "venv",
+            str(venv_path),
+            "--python",
+            "3.12",
             stdout=asyncio.subprocess.PIPE,
             stderr=asyncio.subprocess.PIPE,
         )
@@ -280,8 +314,13 @@ async def setup_embedded_environment() -> dict[str, Any]:
 
         # Install pgserver and psycopg2
         proc = await asyncio.create_subprocess_exec(
-            "uv", "pip", "install", "pgserver", "psycopg2-binary",
-            "--python", str(python_exe),
+            "uv",
+            "pip",
+            "install",
+            "pgserver",
+            "psycopg2-binary",
+            "--python",
+            str(python_exe),
             stdout=asyncio.subprocess.PIPE,
             stderr=asyncio.subprocess.PIPE,
         )
diff --git a/opc/scripts/setup/wizard.py b/opc/scripts/setup/wizard.py
index 142ddc78..c4395ed6 100644
--- a/opc/scripts/setup/wizard.py
+++ b/opc/scripts/setup/wizard.py
@@ -39,6 +39,7 @@
     console = Console()
 except ImportError:
     rich_escape = lambda x: x  # No escaping needed without Rich
+
     # Fallback for minimal environments
     class Console:
         def print(self, *args, **kwargs):
@@ -125,6 +126,19 @@ async def check_container_runtime() -> dict[str, Any]:
             - version: str | None - Version string
             - daemon_running: bool - True if service is responding
     """
+    import socket
+
+    def is_port_in_use(port: int, host: str = "localhost") -> bool:
+        """Check if a port is already in use."""
+        sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
+        sock.settimeout(1)
+        try:
+            result = sock.connect_ex((host, port))
+            sock.close()
+            return result == 0
+        except (socket.timeout, OSError):
+            return False
+
     # Try Docker first (most common)
     result = await check_runtime_installed("docker")
     if result["installed"]:
@@ -200,7 +214,9 @@ async def check_prerequisites_with_install_offers() -> dict[str, Any]:
     if not runtime_info["installed"]:
         await offer_docker_install()
     elif not runtime_info.get("daemon_running", False):
-        console.print(f"  [yellow]{runtime_name.title()} is installed but the daemon is not running.[/yellow]")
+        console.print(
+            f"  [yellow]{runtime_name.title()} is installed but the daemon is not running.[/yellow]"
+        )
         if runtime_name == "docker":
             console.print("  Please start Docker Desktop or the Docker service.")
         else:
@@ -209,24 +225,33 @@ async def check_prerequisites_with_install_offers() -> dict[str, Any]:
         # Retry loop for daemon startup
         max_retries = 3
         for attempt in range(max_retries):
-            if Confirm.ask(f"\n  Retry checking {runtime_name} daemon? (attempt {attempt + 1}/{max_retries})", default=True):
+            if Confirm.ask(
+                f"\n  Retry checking {runtime_name} daemon? (attempt {attempt + 1}/{max_retries})",
+                default=True,
+            ):
                 console.print(f"  Checking {runtime_name} daemon...")
                 await asyncio.sleep(2)  # Give daemon time to start
                 runtime_info = await check_runtime_installed(runtime_name)
                 if runtime_info.get("daemon_running", False):
                     result["docker"] = True
                     result["docker_daemon_running"] = True
-                    console.print(f"  [green]OK[/green] {runtime_name.title()} daemon is now running!")
+                    console.print(
+                        f"  [green]OK[/green] {runtime_name.title()} daemon is now running!"
+                    )
                     break
                 else:
-                    console.print(f"  [yellow]{runtime_name.title()} daemon still not running.[/yellow]")
+                    console.print(
+                        f"  [yellow]{runtime_name.title()} daemon still not running.[/yellow]"
+                    )
             else:
                 break
 
     # Check elan/Lean4 (optional, for theorem proving with /prove skill)
     if not result["elan"]:
         console.print("\n  [dim]Optional: Lean4/elan not found (needed for /prove skill)[/dim]")
-        console.print("  [dim]Install with: curl https://raw.githubusercontent.com/leanprover/elan/master/elan-init.sh -sSf | sh[/dim]")
+        console.print(
+            "  [dim]Install with: curl https://raw.githubusercontent.com/leanprover/elan/master/elan-init.sh -sSf | sh[/dim]"
+        )
 
     # elan is optional, so exclude from all_present check
     result["all_present"] = all([result["docker"], result["python"], result["uv"]])
@@ -390,7 +415,9 @@ async def prompt_embedding_config() -> dict[str, str]:
     console.print("    3. openai - OpenAI API (requires API key)")
     console.print("    4. voyage - Voyage AI API (requires API key)")
 
-    provider = Prompt.ask("Embedding provider", choices=["local", "ollama", "openai", "voyage"], default="local")
+    provider = Prompt.ask(
+        "Embedding provider", choices=["local", "ollama", "openai", "voyage"], default="local"
+    )
 
     config = {"provider": provider}
 
@@ -407,7 +434,7 @@ async def prompt_api_keys() -> dict[str, str]:
     """Prompt user for optional API keys.
 
     Returns:
-        dict with keys: perplexity, nia, braintrust
+        dict with keys: perplexity, nia, braintrust, firecrawl, morph
     """
     console.print("\n[bold]API Keys (optional)[/bold]")
     console.print("Press Enter to skip any key you don't have.\n")
@@ -415,11 +442,15 @@ async def prompt_api_keys() -> dict[str, str]:
     perplexity = Prompt.ask("Perplexity API key (web search)", default="")
     nia = Prompt.ask("Nia API key (documentation search)", default="")
     braintrust = Prompt.ask("Braintrust API key (observability)", default="")
+    firecrawl = Prompt.ask("Firecrawl API key (web scraping)", default="")
+    morph = Prompt.ask("Morph API key (data extraction)", default="")
 
     return {
         "perplexity": perplexity,
         "nia": nia,
         "braintrust": braintrust,
+        "firecrawl": firecrawl,
+        "morph": morph,
     }
 
 
@@ -448,11 +479,11 @@ def generate_env_file(config: dict[str, Any], env_path: Path) -> None:
         lines.append(f"# Database Mode: {mode}")
 
         if mode == "docker":
-            host = db.get('host', 'localhost')
-            port = db.get('port', 5432)
-            database = db.get('database', 'continuous_claude')
-            user = db.get('user', 'claude')
-            password = db.get('password', '')
+            host = db.get("host", "localhost")
+            port = db.get("port", 5432)
+            database = db.get("database", "continuous_claude")
+            user = db.get("user", "claude")
+            password = db.get("password", "")
             lines.append(f"POSTGRES_HOST={host}")
             lines.append(f"POSTGRES_PORT={port}")
             lines.append(f"POSTGRES_DB={database}")
@@ -461,7 +492,9 @@ def generate_env_file(config: dict[str, Any], env_path: Path) -> None:
                 lines.append(f"POSTGRES_PASSWORD={password}")
             lines.append("")
             lines.append("# Connection string for scripts (canonical name)")
-            lines.append(f"CONTINUOUS_CLAUDE_DB_URL=postgresql://{user}:{password}@{host}:{port}/{database}")
+            lines.append(
+                f"CONTINUOUS_CLAUDE_DB_URL=postgresql://{user}:{password}@{host}:{port}/{database}"
+            )
         elif mode == "embedded":
             pgdata = db.get("pgdata", "")
             venv = db.get("venv", "")
@@ -500,6 +533,10 @@ def generate_env_file(config: dict[str, Any], env_path: Path) -> None:
                 lines.append(f"NIA_API_KEY={api_keys['nia']}")
             if api_keys.get("braintrust"):
                 lines.append(f"BRAINTRUST_API_KEY={api_keys['braintrust']}")
+            if api_keys.get("firecrawl"):
+                lines.append(f"FIRECRAWL_API_KEY={api_keys['firecrawl']}")
+            if api_keys.get("morph"):
+                lines.append(f"MORPH_API_KEY={api_keys['morph']}")
             lines.append("")
 
     # Write file
@@ -570,15 +607,30 @@ async def run_setup_wizard() -> None:
     console.print("    [bold]docker[/bold]    - PostgreSQL in Docker (recommended)")
     console.print("    [bold]embedded[/bold]  - Embedded PostgreSQL (no Docker needed)")
     console.print("    [bold]sqlite[/bold]    - SQLite fallback (simplest, no cross-terminal)")
-    db_mode = Prompt.ask("\n  Database mode", choices=["docker", "embedded", "sqlite"], default="docker")
+    db_mode = Prompt.ask(
+        "\n  Database mode", choices=["docker", "embedded", "sqlite"], default="docker"
+    )
 
     if db_mode == "embedded":
         from scripts.setup.embedded_postgres import setup_embedded_environment
+
         console.print("  Setting up embedded postgres (creates Python 3.12 environment)...")
         embed_result = await setup_embedded_environment()
         if embed_result["success"]:
-            console.print(f"  [green]OK[/green] Embedded environment ready at {embed_result['venv']}")
-            db_config = {"mode": "embedded", "pgdata": str(embed_result["pgdata"]), "venv": str(embed_result["venv"])}
+            console.print(
+                f"  [green]OK[/green] Embedded environment ready at {embed_result['venv']}"
+            )
+            console.print(
+                "  [dim]To start server: {venv}/bin/python -c \"from pgserver import get_server; get_server('{pgdata}')\"".format(
+                    venv=embed_result["venv"], pgdata=embed_result["pgdata"]
+                )
+            )
+            console.print("  [dim]To run schema: psql -f docker/init-schema.sql <uri>")
+            db_config = {
+                "mode": "embedded",
+                "pgdata": str(embed_result["pgdata"]),
+                "venv": str(embed_result["venv"]),
+            }
         else:
             console.print(f"  [red]ERROR[/red] {embed_result.get('error', 'Unknown')}")
             console.print("  Falling back to Docker mode")
@@ -586,10 +638,14 @@ async def run_setup_wizard() -> None:
 
     if db_mode == "sqlite":
         db_config = {"mode": "sqlite"}
-        console.print("  [yellow]Note:[/yellow] Cross-terminal coordination disabled in SQLite mode")
+        console.print(
+            "  [yellow]Note:[/yellow] Cross-terminal coordination disabled in SQLite mode"
+        )
 
     if db_mode == "docker":
-        console.print("  [dim]Customize host/port for containers (podman, nerdctl) or remote postgres.[/dim]")
+        console.print(
+            "  [dim]Customize host/port for containers (podman, nerdctl) or remote postgres.[/dim]"
+        )
         if Confirm.ask("Configure database connection?", default=True):
             db_config = await prompt_database_config()
             password = Prompt.ask("Database password", password=True, default="claude_dev")
@@ -626,19 +682,28 @@ async def run_setup_wizard() -> None:
     console.print(f"  [green]OK[/green] Generated {env_path}")
 
     # Step 5: Container stack (Sandbox Infrastructure)
+    # Skip if using embedded PostgreSQL or SQLite
     runtime = prereqs.get("container_runtime", "docker")
     console.print(f"\n[bold]Step 6/13: Container Stack (Sandbox Infrastructure)[/bold]")
-    console.print("  The sandbox requires PostgreSQL and Redis for:")
-    console.print("  - Agent coordination and scheduling")
-    console.print("  - Build cache and LSP index storage")
-    console.print("  - Real-time agent status")
-    if Confirm.ask(f"Start {runtime} stack (PostgreSQL, Redis)?", default=True):
-        from scripts.setup.docker_setup import run_migrations, set_container_runtime, start_docker_stack, wait_for_services
+
+    if db_mode == "embedded":
+        console.print("  [dim]Skipped - using embedded PostgreSQL (no Docker needed)[/dim]")
+    elif db_mode == "sqlite":
+        console.print("  [dim]Skipped - using SQLite (no Docker needed)[/dim]")
+    elif Confirm.ask(f"Start {runtime} stack (PostgreSQL, Redis)?", default=True):
+        from scripts.setup.docker_setup import (
+            run_migrations,
+            set_container_runtime,
+            start_docker_stack,
+            wait_for_services,
+        )
 
         # Set the detected runtime before starting
         set_container_runtime(runtime)
 
-        console.print(f"  [dim]Starting containers (first run downloads ~500MB, may take a few minutes)...[/dim]")
+        console.print(
+            f"  [dim]Starting containers (first run downloads ~500MB, may take a few minutes)...[/dim]"
+        )
         result = await start_docker_stack(env_file=env_path)
         if result["success"]:
             console.print(f"  [green]OK[/green] {runtime.title()} stack started")
@@ -656,7 +721,11 @@ async def run_setup_wizard() -> None:
 
     # Step 6: Migrations
     console.print("\n[bold]Step 7/13: Database Setup[/bold]")
-    if Confirm.ask("Run database migrations?", default=True):
+    if db_mode == "embedded":
+        console.print("  [dim]Skipped - run manually after starting server[/dim]")
+    elif db_mode == "sqlite":
+        console.print("  [dim]Skipped - SQLite does not need migrations[/dim]")
+    elif Confirm.ask("Run database migrations?", default=True):
         from scripts.setup.docker_setup import run_migrations, set_container_runtime
 
         # Ensure runtime is set (in case step 5 was skipped)
@@ -739,7 +808,12 @@ async def run_setup_wizard() -> None:
                 console.print(f"  [green]OK[/green] Installed {result['installed_skills']} skills")
                 console.print(f"  [green]OK[/green] Installed {result['installed_rules']} rules")
                 console.print(f"  [green]OK[/green] Installed {result['installed_agents']} agents")
-                console.print(f"  [green]OK[/green] Installed {result['installed_servers']} MCP servers")
+                console.print(
+                    f"  [green]OK[/green] Installed {result['installed_servers']} MCP servers"
+                )
+                console.print(
+                    f"  [green]OK[/green] Installed {result['installed_scripts']} scripts"
+                )
                 if result["merged_items"]:
                     console.print(
                         f"  [green]OK[/green] Merged {len(result['merged_items'])} custom items"
@@ -753,7 +827,9 @@ async def run_setup_wizard() -> None:
                     console.print(f"  [green]OK[/green] {build_msg}")
                 else:
                     console.print(f"  [yellow]WARN[/yellow] {build_msg}")
-                    console.print("  [dim]You can build manually: cd ~/.claude/hooks && npm install && npm run build[/dim]")
+                    console.print(
+                        "  [dim]You can build manually: cd ~/.claude/hooks && npm install && npm run build[/dim]"
+                    )
             else:
                 console.print(f"  [red]ERROR[/red] {result.get('error', 'Unknown error')}")
         elif choice == "3":
@@ -761,9 +837,13 @@ async def run_setup_wizard() -> None:
             result = install_opc_integration_symlink(claude_dir, opc_source)
 
             if result["success"]:
-                console.print(f"  [green]OK[/green] Symlinked: {', '.join(result['symlinked_dirs'])}")
+                console.print(
+                    f"  [green]OK[/green] Symlinked: {', '.join(result['symlinked_dirs'])}"
+                )
                 if result["backed_up_dirs"]:
-                    console.print(f"  [green]OK[/green] Backed up: {', '.join(result['backed_up_dirs'])}")
+                    console.print(
+                        f"  [green]OK[/green] Backed up: {', '.join(result['backed_up_dirs'])}"
+                    )
                 console.print("  [dim]Changes in ~/.claude/ now sync to repo automatically[/dim]")
 
                 # Build TypeScript hooks
@@ -774,7 +854,9 @@ async def run_setup_wizard() -> None:
                     console.print(f"  [green]OK[/green] {build_msg}")
                 else:
                     console.print(f"  [yellow]WARN[/yellow] {build_msg}")
-                    console.print("  [dim]You can build manually: cd ~/.claude/hooks && npm install && npm run build[/dim]")
+                    console.print(
+                        "  [dim]You can build manually: cd ~/.claude/hooks && npm install && npm run build[/dim]"
+                    )
             else:
                 console.print(f"  [red]ERROR[/red] {result.get('error', 'Unknown error')}")
         else:
@@ -801,7 +883,9 @@ async def run_setup_wizard() -> None:
                 console.print(f"  [green]OK[/green] Installed {result['installed_skills']} skills")
                 console.print(f"  [green]OK[/green] Installed {result['installed_rules']} rules")
                 console.print(f"  [green]OK[/green] Installed {result['installed_agents']} agents")
-                console.print(f"  [green]OK[/green] Installed {result['installed_servers']} MCP servers")
+                console.print(
+                    f"  [green]OK[/green] Installed {result['installed_servers']} MCP servers"
+                )
 
                 # Build TypeScript hooks
                 console.print("  Building TypeScript hooks...")
@@ -811,7 +895,9 @@ async def run_setup_wizard() -> None:
                     console.print(f"  [green]OK[/green] {build_msg}")
                 else:
                     console.print(f"  [yellow]WARN[/yellow] {build_msg}")
-                    console.print("  [dim]You can build manually: cd ~/.claude/hooks && npm install && npm run build[/dim]")
+                    console.print(
+                        "  [dim]You can build manually: cd ~/.claude/hooks && npm install && npm run build[/dim]"
+                    )
             else:
                 console.print(f"  [red]ERROR[/red] {result.get('error', 'Unknown error')}")
         elif choice == "2":
@@ -819,7 +905,9 @@ async def run_setup_wizard() -> None:
             result = install_opc_integration_symlink(claude_dir, opc_source)
 
             if result["success"]:
-                console.print(f"  [green]OK[/green] Symlinked: {', '.join(result['symlinked_dirs'])}")
+                console.print(
+                    f"  [green]OK[/green] Symlinked: {', '.join(result['symlinked_dirs'])}"
+                )
                 console.print("  [dim]Changes in ~/.claude/ now sync to repo automatically[/dim]")
 
                 # Build TypeScript hooks
@@ -830,37 +918,73 @@ async def run_setup_wizard() -> None:
                     console.print(f"  [green]OK[/green] {build_msg}")
                 else:
                     console.print(f"  [yellow]WARN[/yellow] {build_msg}")
-                    console.print("  [dim]You can build manually: cd ~/.claude/hooks && npm install && npm run build[/dim]")
+                    console.print(
+                        "  [dim]You can build manually: cd ~/.claude/hooks && npm install && npm run build[/dim]"
+                    )
             else:
                 console.print(f"  [red]ERROR[/red] {result.get('error', 'Unknown error')}")
         else:
             console.print("  Skipped integration installation")
 
-    # Set CLAUDE_OPC_DIR environment variable for skills to find scripts
-    console.print("  Setting CLAUDE_OPC_DIR environment variable...")
+    # Set CLAUDE_OPC_DIR and CLAUDE_CC_DIR environment variables
+    # CLAUDE_OPC_DIR = opc/ directory (Python scripts, MCP runtime)
+    # CLAUDE_CC_DIR  = repository root (contains .claude/scripts/, .claude/skills/)
+    console.print("  Setting environment variables...")
     shell_config = None
     shell = os.environ.get("SHELL", "")
     if "zsh" in shell:
         shell_config = Path.home() / ".zshrc"
     elif "bash" in shell:
         shell_config = Path.home() / ".bashrc"
+    elif "fish" in shell:
+        fish_config_dir = Path.home() / ".config" / "fish"
+        fish_config_dir.mkdir(parents=True, exist_ok=True)
+        shell_config = fish_config_dir / "config.fish"
 
     opc_dir = _project_root  # Use script location, not cwd (robust if invoked from elsewhere)
-    if shell_config and shell_config.exists():
-        content = shell_config.read_text()
-        export_line = f'export CLAUDE_OPC_DIR="{opc_dir}"'
+    cc_dir = opc_dir.parent  # Repository root (Continuous-Claude-v3/)
+
+    if shell_config and (shell_config.exists() or "fish" in shell):
+        if shell_config.exists():
+            content = shell_config.read_text()
+        else:
+            content = ""
+        changed = False
+        export_opc = None
+        export_cc = None
+
+        if "zsh" in shell or "bash" in shell:
+            export_opc = f'export CLAUDE_OPC_DIR="{opc_dir}"'
+            export_cc = f'export CLAUDE_CC_DIR="{cc_dir}"'
+        elif "fish" in shell:
+            export_opc = f'set -gx CLAUDE_OPC_DIR "{opc_dir}"'
+            export_cc = f'set -gx CLAUDE_CC_DIR "{cc_dir}"'
+
         if "CLAUDE_OPC_DIR" not in content:
             with open(shell_config, "a") as f:
-                f.write(f"\n# Continuous-Claude OPC directory (for skills to find scripts)\n{export_line}\n")
-            console.print(f"  [green]OK[/green] Added CLAUDE_OPC_DIR to {shell_config.name}")
+                f.write(f"\n# Continuous-Claude directories\n{export_opc}\n{export_cc}\n")
+            changed = True
+        elif "CLAUDE_CC_DIR" not in content:
+            with open(shell_config, "a") as f:
+                f.write(f"\n# Continuous-Claude root directory\n{export_cc}\n")
+            changed = True
+
+        if changed:
+            console.print(
+                f"  [green]OK[/green] Added CLAUDE_OPC_DIR and CLAUDE_CC_DIR to {shell_config.name}"
+            )
         else:
-            console.print(f"  [dim]CLAUDE_OPC_DIR already in {shell_config.name}[/dim]")
+            console.print(
+                f"  [dim]CLAUDE_OPC_DIR and CLAUDE_CC_DIR already in {shell_config.name}[/dim]"
+            )
     elif sys.platform == "win32":
         console.print("  [yellow]NOTE[/yellow] Add to your environment:")
         console.print(f'       set CLAUDE_OPC_DIR="{opc_dir}"')
+        console.print(f'       set CLAUDE_CC_DIR="{cc_dir}"')
     else:
         console.print("  [yellow]NOTE[/yellow] Add to your shell config:")
         console.print(f'       export CLAUDE_OPC_DIR="{opc_dir}"')
+        console.print(f'       export CLAUDE_CC_DIR="{cc_dir}"')
 
     # Step 8: Math Features (Optional)
     console.print("\n[bold]Step 9/13: Math Features (Optional)[/bold]")
@@ -921,8 +1045,94 @@ async def run_setup_wizard() -> None:
         console.print("  Skipped math features")
         console.print("  [dim]Install later with: uv sync --extra math[/dim]")
 
-    # Step 9: TLDR Code Analysis Tool
-    console.print("\n[bold]Step 10/13: TLDR Code Analysis Tool[/bold]")
+    # Step 11: qlty CLI (Universal Code Quality)
+    console.print("\n[bold]Step 11/15: qlty Code Quality Tool[/bold]")
+    console.print(
+        "  qlty is a universal code quality tool supporting 70+ linters for 40+ languages."
+    )
+    console.print("  Unlocks: qlty-check, qlty-during-development, fix (deps scope) skills.")
+    console.print("")
+    console.print("  [dim]Free and open source - no API key needed.[/dim]")
+
+    # Check if qlty is already installed
+    if shutil.which("qlty"):
+        console.print("  [green]OK[/green] qlty is already installed")
+    elif Confirm.ask("\nInstall qlty code quality tool?", default=True):
+        console.print("  Installing qlty via curl...")
+        try:
+            result = subprocess.run(
+                ["curl", "-fsSL", "https://qlty.sh/install.sh"],
+                capture_output=True,
+                text=True,
+                timeout=60,
+            )
+            if result.returncode == 0:
+                install_result = subprocess.run(
+                    result.stdout.strip(),
+                    shell=True,
+                    capture_output=True,
+                    text=True,
+                    timeout=120,
+                )
+                if install_result.returncode == 0:
+                    console.print("  [green]OK[/green] qlty installed")
+                else:
+                    console.print("  [yellow]WARN[/yellow] qlty install had issues")
+            else:
+                console.print("  [yellow]WARN[/yellow] Could not install qlty")
+        except subprocess.TimeoutExpired:
+            console.print("  [yellow]WARN[/yellow] Installation timed out")
+        except Exception as e:
+            console.print(f"  [yellow]WARN[/yellow] {e}")
+    else:
+        console.print("  Skipped qlty installation")
+        console.print("  [dim]Install later with: curl -fsSL https://qlty.sh/install.sh | sh[/dim]")
+
+    # Step 12: ast-grep (AST-based Code Search)
+    console.print("\n[bold]Step 12/15: ast-grep Code Analysis Tool[/bold]")
+    console.print("  ast-grep performs AST-based structural code search and refactoring.")
+    console.print("  Unlocks: ast-grep-find, search-router, search-tools skills.")
+    console.print("")
+    console.print("  [dim]Free and open source - no API key needed.[/dim]")
+
+    # Check if ast-grep is already installed
+    if shutil.which("ast-grep"):
+        console.print("  [green]OK[/green] ast-grep is already installed")
+    elif Confirm.ask("\nInstall ast-grep code analysis tool?", default=True):
+        console.print("  Installing ast-grep...")
+        try:
+            result = subprocess.run(
+                ["cargo", "install", "ast-grep"],
+                capture_output=True,
+                text=True,
+                timeout=600,
+            )
+            if result.returncode == 0:
+                console.print("  [green]OK[/green] ast-grep installed via cargo")
+            else:
+                result = subprocess.run(
+                    ["npm", "install", "-g", "ast-grep"],
+                    capture_output=True,
+                    text=True,
+                    timeout=120,
+                )
+                if result.returncode == 0:
+                    console.print("  [green]OK[/green] ast-grep installed via npm")
+                else:
+                    console.print("  [yellow]WARN[/yellow] Could not install ast-grep")
+                    console.print("  Try: cargo install ast-grep")
+        except subprocess.TimeoutExpired:
+            console.print(
+                "  [yellow]WARN[/yellow] Installation timed out (cargo builds from source)"
+            )
+        except Exception as e:
+            console.print(f"  [yellow]WARN[/yellow] {e}")
+    else:
+        console.print("  Skipped ast-grep installation")
+        console.print("  [dim]Install later with: cargo install ast-grep[/dim]")
+
+    # Step 13: TLDR Code Analysis Tool
+    console.print("\n[bold]Step 13/15: TLDR Code Analysis Tool[/bold]")
     console.print("  TLDR provides token-efficient code analysis for LLMs:")
     console.print("    - 95% token savings vs reading raw files")
     console.print("    - 155x faster queries with daemon mode")
@@ -931,7 +1141,21 @@ async def run_setup_wizard() -> None:
     console.print("")
     console.print("  [dim]Note: First semantic search downloads ~1.3GB embedding model.[/dim]")
 
-    if Confirm.ask("\nInstall TLDR code analysis tool?", default=True):
+    # Check if tldr is already installed (llm-tldr, not tldr-pages)
+    is_llm_tldr = False
+    tldr_check = shutil.which("tldr")
+    if tldr_check:
+        verify_result = subprocess.run(
+            [tldr_check, "--help"],
+            capture_output=True,
+            text=True,
+            timeout=10,
+        )
+        is_llm_tldr = any(cmd in verify_result.stdout for cmd in ["tree", "structure", "daemon"])
+
+    if is_llm_tldr:
+        console.print("  [green]OK[/green] TLDR is already installed")
+    elif Confirm.ask("\nInstall TLDR code analysis tool?", default=True):
         console.print("  Installing TLDR...")
         import subprocess
 
@@ -957,13 +1181,21 @@ async def run_setup_wizard() -> None:
                     timeout=10,
                 )
                 # Check if this is llm-tldr (has 'tree', 'structure', 'daemon') not tldr-pages
-                is_llm_tldr = any(cmd in verify_result.stdout for cmd in ["tree", "structure", "daemon"])
+                is_llm_tldr = any(
+                    cmd in verify_result.stdout for cmd in ["tree", "structure", "daemon"]
+                )
                 if verify_result.returncode == 0 and is_llm_tldr:
                     console.print("  [green]OK[/green] TLDR CLI available")
                 elif verify_result.returncode == 0 and not is_llm_tldr:
-                    console.print("  [yellow]WARN[/yellow] Wrong tldr detected (tldr-pages, not llm-tldr)")
-                    console.print("  [yellow]    [/yellow] The 'tldr' command is shadowed by tldr-pages.")
-                    console.print("  [yellow]    [/yellow] Uninstall tldr-pages: pip uninstall tldr")
+                    console.print(
+                        "  [yellow]WARN[/yellow] Wrong tldr detected (tldr-pages, not llm-tldr)"
+                    )
+                    console.print(
+                        "  [yellow]    [/yellow] The 'tldr' command is shadowed by tldr-pages."
+                    )
+                    console.print(
+                        "  [yellow]    [/yellow] Uninstall tldr-pages: pip uninstall tldr"
+                    )
                     console.print("  [yellow]    [/yellow] Or use full path: ~/.local/bin/tldr")
 
                 if is_llm_tldr:
@@ -977,14 +1209,15 @@ async def run_setup_wizard() -> None:
                     console.print("")
                     console.print("  [bold]Semantic Search Configuration[/bold]")
                     console.print("  Natural language code search using AI embeddings.")
-                    console.print("  [dim]First run downloads ~1.3GB model and indexes your codebase.[/dim]")
+                    console.print(
+                        "  [dim]First run downloads ~1.3GB model and indexes your codebase.[/dim]"
+                    )
                     console.print("  [dim]Auto-reindexes in background when files change.[/dim]")
 
                     if Confirm.ask("\n  Enable semantic search?", default=True):
                         # Get threshold
                         threshold_str = Prompt.ask(
-                            "  Auto-reindex after how many file changes?",
-                            default="20"
+                            "  Auto-reindex after how many file changes?", default="20"
                         )
                         try:
                             threshold = int(threshold_str)
@@ -1005,6 +1238,7 @@ async def run_setup_wizard() -> None:
                         has_gpu = False
                         try:
                             import torch
+
                             has_gpu = torch.cuda.is_available() or torch.backends.mps.is_available()
                         except ImportError:
                             pass  # No torch = assume no GPU
@@ -1025,7 +1259,9 @@ async def run_setup_wizard() -> None:
 
                         settings_path.parent.mkdir(parents=True, exist_ok=True)
                         settings_path.write_text(json.dumps(settings, indent=2))
-                        console.print(f"  [green]OK[/green] Semantic search enabled (threshold: {threshold})")
+                        console.print(
+                            f"  [green]OK[/green] Semantic search enabled (threshold: {threshold})"
+                        )
 
                         # Offer to pre-download embedding model
                         # Note: We only download the model here, not index any directory.
@@ -1035,7 +1271,13 @@ async def run_setup_wizard() -> None:
                             try:
                                 # Just load the model to trigger download (no indexing)
                                 download_result = subprocess.run(
-                                    [sys.executable, "-c", f"from tldr.semantic import get_model; get_model('{model}')"],
+                                    [
+                                        "uv",
+                                        "run",
+                                        "python",
+                                        "-c",
+                                        f"from tldr.semantic import get_model; get_model('{model}')",
+                                    ],
                                     capture_output=True,
                                     text=True,
                                     timeout=timeout,
@@ -1052,7 +1294,9 @@ async def run_setup_wizard() -> None:
                             except Exception as e:
                                 console.print(f"  [yellow]WARN[/yellow] {e}")
                         else:
-                            console.print("  [dim]Model downloads on first use of: tldr semantic index .[/dim]")
+                            console.print(
+                                "  [dim]Model downloads on first use of: tldr semantic index .[/dim]"
+                            )
                     else:
                         console.print("  Semantic search disabled")
                         console.print("  [dim]Enable later in .claude/settings.json[/dim]")
@@ -1073,15 +1317,19 @@ async def run_setup_wizard() -> None:
         console.print("  [dim]Install later with: uv tool install llm-tldr[/dim]")
 
         # Ask to disable hooks since they are pre-configured in settings.json
-        if Confirm.ask("\n  Disable TLDR hooks in settings.json? (Avoids crashes if TLDR missing)", default=False):
+        if Confirm.ask(
+            "\n  Disable TLDR hooks in settings.json? (Avoids crashes if TLDR missing)",
+            default=False,
+        ):
             settings_path = get_global_claude_dir() / "settings.json"
             if settings_path.exists():
                 from scripts.setup.claude_integration import strip_tldr_hooks_from_settings
+
                 if strip_tldr_hooks_from_settings(settings_path):
                     console.print("  [green]OK[/green] TLDR hooks disabled")
 
     # Step 10: Diagnostics Tools (Shift-Left Feedback)
-    console.print("\n[bold]Step 11/13: Diagnostics Tools (Shift-Left Feedback)[/bold]")
+    console.print("\n[bold]Step 14/15: Diagnostics Tools (Shift-Left Feedback)[/bold]")
     console.print("  Claude gets immediate type/lint feedback after editing files.")
     console.print("  This catches errors before tests run (shift-left).")
     console.print("")
@@ -1119,7 +1367,7 @@ async def run_setup_wizard() -> None:
     console.print("  [dim]TypeScript, Go, Rust coming soon.[/dim]")
 
     # Step 11: Loogle (Lean 4 type search for /prove skill)
-    console.print("\n[bold]Step 12/13: Loogle (Lean 4 Type Search)[/bold]")
+    console.print("\n[bold]Step 15/15: Loogle (Lean 4 Type Search)[/bold]")
     console.print("  Loogle enables type-aware search of Mathlib theorems:")
     console.print("    - Used by /prove skill for theorem proving")
     console.print("    - Search by type signature (e.g., 'Nontrivial _ ↔ _')")
@@ -1133,7 +1381,9 @@ async def run_setup_wizard() -> None:
         # Check elan prerequisite
         if not shutil.which("elan"):
             console.print("  [yellow]WARN[/yellow] Lean 4 (elan) not installed")
-            console.print("  Install with: curl https://raw.githubusercontent.com/leanprover/elan/master/elan-init.sh -sSf | sh")
+            console.print(
+                "  Install with: curl https://raw.githubusercontent.com/leanprover/elan/master/elan-init.sh -sSf | sh"
+            )
             console.print("  Then re-run the wizard to install Loogle.")
         else:
             console.print("  [green]OK[/green] elan found")
@@ -1161,7 +1411,9 @@ async def run_setup_wizard() -> None:
                     if result.returncode == 0:
                         console.print("  [green]OK[/green] Updated")
                     else:
-                        console.print(f"  [yellow]WARN[/yellow] Update failed: {result.stderr[:100]}")
+                        console.print(
+                            f"  [yellow]WARN[/yellow] Update failed: {result.stderr[:100]}"
+                        )
             else:
                 console.print(f"  Cloning Loogle to {loogle_home}...")
                 loogle_home.parent.mkdir(parents=True, exist_ok=True)
@@ -1170,7 +1422,7 @@ async def run_setup_wizard() -> None:
                         ["git", "clone", "https://github.com/nomeata/loogle", str(loogle_home)],
                         capture_output=True,
                         text=True,
-                        timeout=120,
+                        timeout=600,  # 10 min for 2GB repo
                     )
                     if result.returncode == 0:
                         console.print("  [green]OK[/green] Cloned")
@@ -1198,10 +1450,16 @@ async def run_setup_wizard() -> None:
                     else:
                         console.print(f"  [red]ERROR[/red] Build failed")
                         console.print(f"       {result.stderr[:200]}")
-                        console.print("  You can build manually: cd ~/.local/share/loogle && lake build")
+                        console.print(
+                            "  You can build manually: cd ~/.local/share/loogle && lake build"
+                        )
                 except subprocess.TimeoutExpired:
-                    console.print("  [yellow]WARN[/yellow] Build timed out (this is normal for first build)")
-                    console.print("  Continue building manually: cd ~/.local/share/loogle && lake build")
+                    console.print(
+                        "  [yellow]WARN[/yellow] Build timed out (this is normal for first build)"
+                    )
+                    console.print(
+                        "  Continue building manually: cd ~/.local/share/loogle && lake build"
+                    )
                 except Exception as e:
                     console.print(f"  [red]ERROR[/red] {e}")
 
@@ -1254,7 +1512,7 @@ async def run_setup_wizard() -> None:
                 console.print(f"  [yellow]WARN[/yellow] loogle_search.py not found at {src_script}")
 
             console.print("")
-            console.print("  [dim]Usage: loogle-search \"Nontrivial _ ↔ _\"[/dim]")
+            console.print('  [dim]Usage: loogle-search "Nontrivial _ ↔ _"[/dim]')
             console.print("  [dim]Or use /prove skill which calls it automatically[/dim]")
     else:
         console.print("  Skipped Loogle installation")
@@ -1344,9 +1602,7 @@ async def main():
 
     # Show menu if no args
     if len(sys.argv) == 1:
-        console.print(
-            Panel.fit("[bold]CLAUDE CONTINUITY KIT v3[/bold]", border_style="blue")
-        )
+        console.print(Panel.fit("[bold]CLAUDE CONTINUITY KIT v3[/bold]", border_style="blue"))
         console.print("\n[bold]Options:[/bold]")
         console.print("  [bold]1[/bold] - Install / Update")
         console.print("  [bold]2[/bold] - Uninstall (restore backup)")