Topic hub
Agent memory
Evidence on agent memory as a write loop: what should persist, who can correct it, and how it changes future work.
Retrieval answer
Evidence on agent memory as a write loop: what should persist, who can correct it, and how it changes future work. Agent memory is not only retrieval; it is a governed write loop that decides what survives a task. Useful memory needs correction-time updates, provenance, decay, and behavioral evaluation.
Pattern memory
What patterns are emerging?
- high
Company memory needs write and correction loops
Company memory becomes useful when real work continuously updates a governed knowledge layer, not when an agent merely searches a larger document pile.
Field notes
What should readers understand next?
ABBEL Treats Memory as an Explicit Belief State
The BAIR ABBEL post reframes long-horizon memory as a learned natural-language belief state rather than raw context accumulation.
Mem0 Shows Agent Memory Should Not All Live in Context
Mem0's Claude Code experiment separates durable memory from the conversation window: retrieve the relevant slice, survive /clear, and avoid loading every memory file up front.
Hermes + Obsidian Turns Notes Into an Agent-Operable Workspace
Hermes now ships an Obsidian skill that can read, search, create, edit, and link vault notes, making a filesystem knowledge base actionable agent context.
Claude Cowork Turns Screen Recordings Into Reusable Skills
Claude Cowork's recorded-skill flow makes workflow capture a first-party path from human demonstration to reusable agent capability.
Hermes Agent Makes the Learning Loop Part of the Runtime
Nous Research's Hermes Agent packages memory, skills, messaging, scheduling, tool use, and sandboxing as a runtime rather than a single chat surface.
Cognee Packages Agent Memory as a Self-Hosted Knowledge Graph
Cognee matters less as another memory SDK and more as an attempt to package persistent agent memory through ingestion, graph/vector search, ontology, and self-hosting.
Raw signals
What changed recently?
mem0: Experiment 2: Does a preference survive a hard context wipe?
A public X post from mem0 with a linked primary source flags Experiment 2: Does a preference survive a hard context wipe? We told Claude(mid-conversation) to write functions with standard for loops instead of list comprehensions. Then ran /...
Letta: Access stateful agents through any OpenAI-compatible interface.
A public X post from Letta as a public source in its own right flags Access stateful agents through any OpenAI-compatible interface. Run `letta server --openai-api` to chat with stateful agents through @OpenWebUI @LibreChatAI @jandotai & more.
Letta: Opus 5 is now available in Letta!
A public X post from Letta as a public source in its own right flags Opus 5 is now available in Letta! In our internal evaluations, we found that Opus 5: - Matches Fable in memory generation (rewriting memory and learning skills) - Significantly im...
LangChain: We’re hosting a meetup with @cognition on Tuesday.
A public X post from LangChain with a linked primary source flags We’re hosting a meetup with @cognition on Tuesday. Come learn about LLM Wikis and how to give your agents open memory with OpenWiki. RSVP while spaces last:
mem0: 5/ Myth: Memory is only for chatbots Coding agents forget your codebase conventions every sess...
A public X post from mem0 as a public source in its own right flags 5/ Myth: Memory is only for chatbots Coding agents forget your codebase conventions every session. Personal assistants forget your calendar preferences by Tuesday. Any agent that...
Letta: Office Hours are today!
A public X post from Letta with a linked primary source flags Office Hours are today! Come join the livestream at 11:30am Pacific. New stuff: - Cloud sandboxes are now persistent -- one per agent - GitHub integration -- clone repositories in...
Letta: Letta agents now come with built-in image generation from @bfl_ai @GeminiApp @OpenAI Ask your...
A public X post from Letta with a linked primary source flags Letta agents now come with built-in image generation from @bfl_ai @GeminiApp @OpenAI Ask your agent to generate or modify images to try it out!
Agent memory is being evaluated as system behavior
MemoryData evaluates what an agent stores, retrieves, updates, and forgets across a sequence instead of grading one final answer.
Codebase Memory Combines Search with a Repository Map
A code retrieval tool adds structural repository context to text search so an agent can locate relevant code without loading the whole project into context.
Agent Memory Needs Its Own Review Loop
Persistent memory can amplify mistakes across future tasks, so edits, provenance, and correction paths need explicit operator control.
Deep Agents combines plans, files, memory, and subagents
The open-source Deep Agents stack packages common long-horizon primitives on top of LangGraph and LangChain.
DoorDash makes memory a versioned personalization layer
DoorDash builds structured memory blocks from behavior and stores model, schema, timestamp, prompt, and response lineage for each component.
OpenHuman Builds a Personal Harness Around Local Memory
OpenHuman combines personal data, memory, tools, and an operator layer into a durable system instead of another isolated assistant chat.
GitHub / yichuan-w/LEANN: Routable Model Components
The archive captures GitHub / yichuan-w/LEANN as a dated public record from GitHub / yichuan-w/LEANN. It documents models becoming replaceable or specialized components inside a more durable runtime and is retained as supporting evidence for the routable model components trend.
AgentMemory Shares State Across Codex, Claude Code, and Cursor
A common memory layer reduces the cost of switching coding clients and moves context ownership outside any one model interface.
OpenAI Shows Memory Compaction for Long-Running Agents
The Agents SDK pattern treats compaction and state transfer as explicit reliability mechanisms for work that outlives one context window.
CocoIndex Keeps Agent Context Fresh Without Full Reindexing
CocoIndex turns source changes into incremental dataflow updates for live agent context and code-aware retrieval.
Gemini CLI: Gemini CLI v0.40.0 Release Notes📝 • New tiered memory system 🧠 • Automatic skills generated ba...
A public X post from Gemini CLI as a public source in its own right flags Gemini CLI v0.40.0 Release Notes📝 • New tiered memory system 🧠 • Automatic skills generated based on past sessions 🛠️ • Gemma support for local routing 💎 • Streamlined UI with com...
GitHub / chenglou/pretext: Executable Design Context
The archive captures GitHub / chenglou/pretext as a dated public record from GitHub / chenglou/pretext. It documents design systems becoming machine-readable context, constraints, and review loops and is retained as supporting evidence for the executable design context trend.
Hugging Face: Verification Bandwidth
The archive captures Hugging Face as a dated public record from Hugging Face. It documents evaluation, review, and observability becoming the bottleneck after generation accelerates and is retained as pressure-testing evidence for the verification bandwidth trend.
Google Research / Building Better Ai Benchmarks How Many Raters: Verification Bandwidth
The archive captures Google Research / Building Better Ai Benchmarks How Many Raters as a dated public record from Google Research / Building Better Ai Benchmarks How Many Raters. It documents evaluation, review, and observability becoming the bottleneck after generation accelerates and is retained as pressure-testing evidence for the verification bandwidth trend.
Mckinsey / Rethinking Enterprise Architecture For Agentic Era: Ai-Native Operating Models
The archive captures Mckinsey / Rethinking Enterprise Architecture For Agentic Era as a dated public record from Mckinsey / Rethinking Enterprise Architecture For Agentic Era. It documents AI adoption shifting jobs, coordination, review, and organizational capacity and is retained as pressure-testing evidence for the AI-native operating models trend.
OpenAI Developers / Designing Delightful Frontends With Gpt: Executable Design Context
The archive captures OpenAI Developers / Designing Delightful Frontends With Gpt as a dated public record from OpenAI Developers / Designing Delightful Frontends With Gpt. It documents design systems becoming machine-readable context, constraints, and review loops and is retained as supporting evidence for the executable design context trend.
GitHub / Agent-Field/agentfield: Agent-Ready Software
The archive captures GitHub / Agent-Field/agentfield as a dated public record from GitHub / Agent-Field/agentfield. It documents software exposing explicit capabilities, permissions, and machine-readable actions and is retained as pressure-testing evidence for the agent-ready software trend.