-
Notifications
You must be signed in to change notification settings - Fork 4
Issues
is:issue state:open
is:issue state:open
Issue creation is restricted in this repository
Search results
feat(session): implement session persistence, event log replay, and zeph serve mode
enhancementNew feature or requestNew feature or requestStatus: Open.#5343 In bug-ops/zeph;feat(cocoon): implement managed sidecar lifecycle (cocoon_managed config + TaskSupervisor)
enhancementNew feature or requestNew feature or requestStatus: Open.#5339 In bug-ops/zeph;feat(cocoon): implement E2E payload encryption (Option A, client-side Ed25519)
enhancementNew feature or requestNew feature or requestStatus: Open.#5338 In bug-ops/zeph;chore(acp): track ACP v2 breaking changes (prompt lifecycle, remote transports, capabilities)
P4Long-term / exploratoryLong-term / exploratoryresearchResearch-driven improvementResearch-driven improvementStatus: Open.#4461 In bug-ops/zeph;feat(acp): track MCP-over-ACP transport (unstable)
P4Long-term / exploratoryLong-term / exploratoryresearchResearch-driven improvementResearch-driven improvementStatus: Open.#4460 In bug-ops/zeph;research: Scaling Teams or Scaling Time — lifelong learning via episodic/procedural/transactive memory (2604.03295)
memoryzeph-memory crate (SQLite)zeph-memory crate (SQLite)P4Long-term / exploratoryLong-term / exploratoryresearchResearch-driven improvementResearch-driven improvementStatus: Open.#4209 In bug-ops/zeph;research: LLM serving mathematical optimization — request routing & cache scheduling (2605.01280)
llmzeph-llm crate (Ollama, Claude)zeph-llm crate (Ollama, Claude)P4Long-term / exploratoryLong-term / exploratoryresearchResearch-driven improvementResearch-driven improvementStatus: Open.#4208 In bug-ops/zeph;research: Joint Optimization of Multi-agent Memory System (2605.12631)
memoryzeph-memory crate (SQLite)zeph-memory crate (SQLite)P4Long-term / exploratoryLong-term / exploratoryresearchResearch-driven improvementResearch-driven improvementStatus: Open.#4207 In bug-ops/zeph;research: GroupMemBench — benchmarking LLM agent memory in multi-party conversations
llmzeph-llm crate (Ollama, Claude)zeph-llm crate (Ollama, Claude)memoryzeph-memory crate (SQLite)zeph-memory crate (SQLite)P4Long-term / exploratoryLong-term / exploratoryresearchResearch-driven improvementResearch-driven improvementStatus: Open.#4206 In bug-ops/zeph;research(core): ExpThink experience-guided RL with self-evolving curriculum for zeph-agent-feedback (arXiv:2605.07501)
P4Long-term / exploratoryLong-term / exploratoryresearchResearch-driven improvementResearch-driven improvementStatus: Open.#4201 In bug-ops/zeph;research(memory): DeMem rate-distortion framework for decision-centric agent memory — optimal forgetting boundary (arXiv:2605.10870)
memoryzeph-memory crate (SQLite)zeph-memory crate (SQLite)P4Long-term / exploratoryLong-term / exploratoryresearchResearch-driven improvementResearch-driven improvementStatus: Open.#4200 In bug-ops/zeph;research(bench): multi-agent scientific AI evaluation framework — reasoning vs retrieval separation and contamination-safe ground truth (arXiv:2603.26718)
P4Long-term / exploratoryLong-term / exploratoryresearchResearch-driven improvementResearch-driven improvementStatus: Open.#4196 In bug-ops/zeph;