Topic
Source-linked workflows that name agent-memory. Part of the AI agent use cases on UsecaseBoard, or browse all use cases.
Open-source local procedural memory that learns only after tests pass, then hands Claude Code/Codex/Cursor a short verified skill note on similar tasks (reported token/tool cuts in published benchmarks).
Tencent Hunyuan open-weights Hy-Memory-30B-A3B (~3B active MoE) to replace the general LLM on the write path of Mem0/MemOS/Hy-Memory—trained with on-policy distillation + attributed memory RL; beats Claude Opus 5 / DeepSeek V4.1 Flash on a fixed cross-harness memory-writer matrix; Apache-friendly HF weights + GitHub framework.
JJ Englert changed his Meta Muse routines to check live sources and review corrections after an obsolete note triggered a reminder for a canceled meeting.
OpenAI's data-analyst example combines schema discovery, metric context, read-only SQL, and explicitly saved corrections in one persistent investigation; database execution stays in the application.
Artem Zhutov used Claude Code to turn recurring corrections from past sessions into an HTML report of proposed instruction changes.
Make's memory guide provides a concrete read-normalize-reason-route-write pattern keyed to a customer or case, with module-level inspection and recovery paths for missing or malformed state.
A Technically writer documents a month-long Claude Code experiment managing a Meta Ads account: fresh sessions read prior logs, pull performance data, decide, act or abstain, and commit the updated history.
A founder describes a working beta that scans a repository, asks product questions, and writes a behavior contract distinguishing confirmed decisions from unsettled assumptions.