merge: AgentProject membership — the junction is now load-bearing

Merges feat/agent-project-membership (task 166/152). An agent can finally join
many projects: membership routes, per-membership role and optional brain
override, chat resolving context from the membership, and CLI attach-agent /
detach-agent / get memberships. The legacy Agent.projectId fallback is kept for
this release as the schema comment promises.

Conflict resolved in main.ts: that branch added its RBAC case to
mapUrlToPermission inline, while this branch had extracted the function to
middleware/url-permission.ts so it could be tested. Kept the extraction and
ported the case into it — joining a team is `edit:agents:<name>`, not
`create:agents`, which is the authority to mint a whole agent.

Also fixes something the extraction makes possible. security.test.ts contained a
hand-copied reproduction of mapUrlToPermission, because main.ts self-invokes
main() and nothing in it was importable. A security test asserting against a
copy stops testing the system the moment the real one changes, and this copy had
already drifted. It now imports the real function; all 18 assertions still pass,
against the code that actually runs.

Local: build clean, workspace 2550 passed, lint unchanged at 794 errors.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4wNHWf7xSwnZCWpJcyv9p

.taskmaster/tasks/task_074.md

@@ -0,0 +1,57 @@
+# Task ID: 74
+
+**Title:** Implement Content Type Detection
+
+**Status:** done
+
+**Dependencies:** None
+
+**Priority:** high
+
+**Description:** Create a utility that detects content type (JSON, YAML, XML, code, prose) for structural splitting in the section-split stage.
+
+**Details:**
+
+Create `src/mcplocal/src/proxymodel/content-detection.ts`:
+
+```typescript
+export type ContentType = 'json' | 'yaml' | 'xml' | 'code' | 'prose';
+
+export function detectContentType(content: string): ContentType {
+  const trimmed = content.trimStart();
+  
+  // JSON detection
+  if (trimmed.startsWith('{') || trimmed.startsWith('[')) {
+    try {
+      JSON.parse(content);
+      return 'json';
+    } catch { /* not valid JSON, continue */ }
+  }
+  
+  // XML detection
+  if (trimmed.startsWith('<?xml') || /^<[a-zA-Z][^>]*>/.test(trimmed)) {
+    return 'xml';
+  }
+  
+  // YAML detection (key: value at start of lines)
+  if (/^[a-zA-Z_][a-zA-Z0-9_]*:\s/m.test(trimmed) && !trimmed.includes('{')) {
+    return 'yaml';
+  }
+  
+  // Code detection (common patterns)
+  const codePatterns = [
+    /^(function |class |def |const |let |var |import |export |package |pub fn |fn |impl )/m,
+    /^#include\s+[<"]/m,
+    /^(public |private |protected )?(static )?(void |int |string |bool )/m,
+  ];
+  if (codePatterns.some(p => p.test(trimmed))) {
+    return 'code';
+  }
+  
+  return 'prose';
+}
+```
+
+**Test Strategy:**
+
+Unit tests with sample content for each type: valid JSON objects/arrays, XML documents, YAML configs, code snippets in multiple languages (JS, Python, Rust, Go, Java), and prose markdown. Edge cases: JSON-like strings that aren't valid JSON, mixed content.