Repository Issue Activity (beta)

#1269 fix: summarizer 模型没有进行“总结”，而是进行“对话”。换了许多个模型仍然一样

#1273 bug: Thread leak in searcher._retrieve_paths causes container thread exhaustion

#1351 feat:

#1448 fix: 添加包含偏好信息的记忆后，使用搜索接口POST /product/search无法检索到任何记忆

#2188 feat: Add native OpenCode local plugin support

#2202 /product REST surface is unauthenticated by default; user_id is client-supplied (cross-user IDOR) and doc_path reads arbitrary server files

#2340 memmy-agent 启动成功后健康检查持续返回 503

#2205 Cypher injection via client-controlled filter keys on /product/search (neo4j)

#2308 LLM facade drops opts.op before providers — per-op provider switches (e.g. thinking kill-switch for summarize) can never fire

#2335 l3.abstraction hard-fails entire world-model generation on empty LLM title (no retry/fallback)

#2336 Feedback classifier leaks raw LLM completion (think tags, special tokens) into persisted feedback rationale

#2333 [Feature Request] memos-local-plugin 完整模式（自进化）支持「闲时批量执行」与 LLM 并发隔离

#2331 删除记忆接口返回成功但记忆未实际删除

#2329 add_memory 返回成功但立即查询同一 memory_id 404

#2327 memmy-agent 第二条消息固定报错 session not found

#1383 [Bug] Multi-Agent mode: cross-agent memory injection + recursive chunk duplication (65908 records in 1 week)

#2170 Request for MSCE paper reproduction artifacts and license clarification

#2319 fix: skill crystallization retries the same failing policies forever

#2318 fix: L2 policy induction crystallizes conversational acts into dead skills

#2320 fix: truncated JSON completions (finish_reason=length) are retried at the same max_tokens

#2317 feat: 建议开发 deepseek-harness 插件

#2313 Merged KV caches report get_seq_length() == 0 and are silently discarded on transformers >= 4.57

#2315 KVCacheMemory._concat_caches merges KV fragments without re-rotating them to their new positions

#1390 feat: python版本能否适配sqlite或者嵌入式数据库这种存储

#1391 fix: