feat(rag): harden post-LightRAG runtime

Retire legacy OCR/media/evidence fallbacks, add local-folder event bus and Page Aggregate guards, and archive completed design checklists.

Validation: cargo test -p mnote-web -- --test-threads=1; cargo test --workspace -- --test-threads=1; git diff --check; codegraph sync .; codegraph_status.

rust/crates/mnote-web/src/hermes_tools/knowledge_rag.rs

@@ -130,6 +130,8 @@ fn compact_query_result_for_agent(payload: Value) -> Value {
         "references": references,
         "citations": citations,
         "sourceScope": payload.get("sourceScope").cloned().unwrap_or(Value::Array(Vec::new())),
+        "sourceScopeMode": payload.get("sourceScopeMode").cloned().unwrap_or_else(|| json!("post_filter_mapped_references")),
+        "rawScopeFiltered": payload.get("rawScopeFiltered").cloned().unwrap_or(Value::Bool(false)),
         "rawStatus": payload.pointer("/raw/status").cloned().unwrap_or(Value::Null),
         "rawMessage": payload.pointer("/raw/message").cloned().unwrap_or(Value::Null),
         "rawMetadata": payload.pointer("/raw/metadata").cloned().unwrap_or(Value::Null),
@@ -155,3 +157,43 @@ fn compact_reference_for_agent(reference: &Value) -> Value {
         "citationUrl": reference.get("citationUrl").cloned().unwrap_or(Value::Null),
     })
 }
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn compact_query_result_marks_post_filter_scope_without_raw_chunks() {
+        let payload = json!({
+            "ok": true,
+            "provider": "lightrag",
+            "sourceScope": ["docs/a.md"],
+            "sourceScopeMode": "post_filter_mapped_references",
+            "rawScopeFiltered": false,
+            "raw": {
+                "status": "success",
+                "chunks": [{"content": "raw chunk must not be exposed to agent"}],
+                "metadata": {"count": 1}
+            },
+            "references": [{
+                "sourceRootRelativePath": "docs/a.md",
+                "quote": "scoped quote",
+                "locatorDegraded": true,
+                "citationMarkdown": "[来源定位降级：docs/a.md](/documents/local-md:docs~2Fa.md)"
+            }]
+        });
+
+        let compact = compact_query_result_for_agent(payload);
+        assert_eq!(
+            compact["sourceScopeMode"].as_str(),
+            Some("post_filter_mapped_references")
+        );
+        assert_eq!(compact["rawScopeFiltered"].as_bool(), Some(false));
+        assert!(compact.get("raw").is_none());
+        assert!(compact.get("chunks").is_none());
+        assert_eq!(
+            compact["references"][0]["sourceRootRelativePath"].as_str(),
+            Some("docs/a.md")
+        );
+    }
+}

rust/crates/mnote-web/src/hermes_tools/manifest.rs

@@ -290,6 +290,7 @@ fn doc_find_tool() -> Value {
     })
 }
 
+// 旧 evidence 工具仅保留为历史对照；当前 manifest() 不注册这些工具，资料库问答走 mnote.knowledge_rag.*。
 #[allow(dead_code)]
 fn evidence_search_tool() -> Value {
     let mut properties = base_identity_properties();
@@ -434,13 +435,13 @@ fn knowledge_rag_query_tool() -> Value {
             json!({
                 "type": "array",
                 "items": { "type": "string" },
-                "description": "可选 MNote workspace 相对路径范围；可传文件或目录，返回 references 会限制在这些来源内。"
+                "description": "可选 MNote workspace 相对路径范围；可传文件或目录。当前语义是 LightRAG provider 检索后，MNote 只过滤返回的 references；provider raw 仍可能是全局结果。"
             }),
         );
     }
     json!({
         "name": "mnote.knowledge_rag.query",
-        "description": "向 LightRAG 资料库提问，返回 provider 原始结果和经 MNote registry 映射后的引用。",
+        "description": "向 LightRAG 资料库提问，返回 provider 原始结果和经 MNote registry 映射、sourcePaths 后过滤的引用。",
         "schemaVersion": TOOL_SCHEMA_VERSION,
         "capabilityScope": ["knowledge_rag.read", "evidence.read"],
         "status": "available",
@@ -477,6 +478,7 @@ fn knowledge_rag_open_reference_tool() -> Value {
     })
 }
 
+// 旧 local index agent 工具仅保留为历史对照；当前 manifest() 不注册这些工具。
 #[allow(dead_code)]
 fn index_status_tool() -> Value {
     let mut properties = base_identity_properties();