fix(proxy): reduce unnecessary Copilot premium interaction consumption

- Fix request classification: treat messages containing tool_result as
  agent continuation instead of user-initiated, preventing false premium
  charges on every tool call
- Add subagent detection via __SUBAGENT_MARKER__ and metadata._agent_
  fallback, setting x-interaction-type=conversation-subagent
- Add deterministic x-interaction-id derived from session ID to group
  requests into a single billing interaction
- Add orphan tool_result sanitization to prevent upstream API errors
  that could cause retries and duplicate billing
- Reorder pipeline: classify (on original body) → sanitize → merge →
  warmup, ensuring classification sees raw tool_result semantics
- Enable warmup downgrade by default with gpt-5-mini model
- Enhance session ID extraction priority chain for Copilot cache keys
- Detect infinite whitespace bug in streaming tool call arguments
This commit is contained in:
Jason
2026-04-14 10:12:45 +08:00
parent c01338ac33
commit 0739b60341
7 changed files with 684 additions and 75 deletions
+1 -1
View File
@@ -337,7 +337,7 @@ fn extract_from_metadata(body: &serde_json::Value) -> Option<SessionIdResult> {
/// 从 user_id 解析 session_id
///
/// 格式: `user_identifier_session_actual_session_id`
fn parse_session_from_user_id(user_id: &str) -> Option<String> {
pub(super) fn parse_session_from_user_id(user_id: &str) -> Option<String> {
// 查找 "_session_" 分隔符
if let Some(pos) = user_id.find("_session_") {
let session_id = &user_id[pos + 9..]; // "_session_" 长度为 9