Turning Agent Memory Into Skills That Work — Will Lyon, Neo4j

Video authored by the channel shown, not by IANIA; all rights belong to the creator. Transcript and summary generated automatically; the original-language content is preserved.

Original language: English

Original video: Turning Agent Memory Into Skills That Work — Will Lyon, Neo4j · Channel: AI Engineer · 18:05 · Watch on YouTube

Topics: AI agents, Data, databases and embeddings

Summary

Will Lyon describes why agent memory needs to capture connected, typed, traceable knowledge—not just retrieve similar text—and presents context graphs combining short-term, long-term, and reasoning memory. He explains how decision traces can record an agent’s actions, evidence, policies, tool calls, and execution details so agents can learn from shared experience. The talk then introduces skill distillation: turning grounded memory graphs into executable, governed skills represented as typed execution graphs. A demo of Neo4j’s agent memory service shows how users can scope, generate, inspect, and package skills while tracking whether their underlying data becomes stale.

Key points

Retrieving relevant text is not enough for useful agent memory; knowledge should be connected, typed, and traceable.
Entity resolution and a shared domain ontology help establish canonical representations of people and other entities.
A context graph can combine short-term, long-term, and reasoning memory.
Reasoning memory captures decision traces, including evidence, policies, execution plans, tool calls, results, token use, and timing.
Shared decision traces can help multiple agents learn from one another.
Skills can be distilled from memory graphs as grounded, deterministic procedural steps rather than only as markdown documents.
APE represents skills as typed execution graphs with schema-governed steps.
Skill quality and governance involve grounding, coverage, coherence, and detecting stale or changed underlying data.
The Neo4j agent memory service provides tools for ingesting, retrieving, exploring, and distilling workspace-scoped memory.

Chapters

Start	Chapter	Summary
00:12	Why agent memory needs more than retrieval	Lyon frames the problem of agents forgetting what they learn and argues for connected, typed, traceable knowledge with canonical entity representations.
03:59	Reasoning memory and shared decision traces	He explains how to record an agent’s reasoning, evidence, policies, tool calls, results, and execution details so other agents can benefit from those traces.
06:07	From memory graphs to executable skills	The talk introduces skills as reusable units and describes APE, which represents skills as typed execution graphs with schema-governed steps.
09:56	Distilling and governing skills	Lyon outlines how to distill grounded, deterministic skills from memory graphs and assess their grounding, coverage, coherence, and freshness.
12:02	The agent memory service and distillation pipeline	He describes the workspace-scoped memory service, its background workers, and a mostly deterministic pipeline that grounds skill components in underlying data.
14:48	Demo: exploring memory and generating a skill	The demo shows the service’s memory tools, healthcare ontology, conversation data, and the process of selecting a scope to distill a skill.
17:29	Grounded skill output and resources	Lyon shows how a patient-intake skill links its steps to tool calls and entities, then closes with slides and project resources.

Quotes

agents need knowledge that is both connected, typed and traceable, right? Not just retrievable. — 01:50

reasoning memory is a first class citizen — 03:59

every decision that our agent makes is linked to the reasoning grounded in evidence — 04:29

We want this to be grounded in actual data that we’ve observed. — 10:27

most of these steps are deterministic. — 13:06

Concepts and entities

Will Lyon, Neo4j, agent memory, Context Graph, reasoning graph, APE, Agent Skills, SkillsBench, NAMS, MCP

Full transcript (original language)

00:12 Let’s go ahead and get started. I see some folks wanting to to come on in. So, feel free to to come in. Uh, this session is going to be all about actionable knowledge and context graphs. So, my name is Will. I’m a product manager at Neo Forj. Um there’s been a lot of discussion in this conference so far about loops, right? The uh React loop, the Ralph loops. Sometimes I feel like we’re in maybe

00:43 like a an amnesic loop, right? where our our agents go through this uh reasoning phase acting. They successfully complete some task but then kind of forget uh what they’ve learned. And memory systems today are largely focused around uh embedding text then at retrieval time sort of finding the most relevant data to to stuff into the context, right? Um so so something like this we find you know similar chunks of data in in our our corpus we inject that into the

01:17 context window and trust that the agent is able to do something useful with that right now the the challenge that we have is you know recall isn’t really like actionable knowledge so one challenge we run into is like for example we don’t have a canonical representation of a thing right so if we have you know three different ways that we refer to Dr. Nwin, Robert N, the cardiologist, right? Like depending on the the context of this discussion, we need to have some

01:50 canonical representation of the thing, right? So really agents need knowledge that is both connected, typed and traceable, right? Not just retrievable. So retrieval is is uh just a part of the problem when we’re talking about uh agent memory. So at NearJ this is how we think of agent memory. We think of it as a connected graph composed of short-term, long-term and reasoning

02:21 memory. Uh we’ll take a look at uh this in a bit more detail. Uh but bear with me on on this idea of context graphs for agent memory. Right? So what are a few of the the key components here? Well, one is going from unstructured data to a knowledge graph, right? Going from the this process of agent messages, right? Both user and assistant messages going through some entity extraction process where we’re identifying what are the entities uh and

02:53 how are they connected. um doing this in a graph where we have strong types, right? We have a a relationship that describes how these entities are related. The entity resol resolution phase is one of the most important parts of building up this this knowledge graph, right? Understanding and making sure that you have a canonical representation of the thing. Uh so when we’re talking about Dr. win we know this is a provider his

03:25 name the role that he has uh and so on and a shared ontology is an important piece of this right so having some description of your data model of the domain that you’re working with this is one of the key pieces uh making sure that you are successful in going from that unstructured text data to uh to a knowledge graph so for folks that have that have worked with memory uh in agents in the past like this this should look a little familiar. Uh one thing though that I think is really

03:59 important when we’re talking about agent memory and actual knowledge is this idea of the reasoning graph right so um if we saw in our our uh sort of three components of agent memory here reasoning memory is a first class citizen right we’ve talked about short-term and long-term these are the messages the entities extracted but what about the actual actions that the agent takes what about the the reasoning um we want to make sure that we’re capturing

04:29 that because that is an important piece as well. We want to make sure that we’re storing thinking the facts not just uh not just the data right so how an agent decided and this is typically represented as decision traces right so every decision uh that our agent makes is linked to uh the reasoning grounded in evidence right we have policies um that that we’re modeling explicitly and we understand for that execution plan the tools the agent called the

05:02 results of those tool calls. Uh we also want to capture things like how many tokens did this burn, the the timing that this took and so on. This is all part of uh capturing the reasoning memory. And this is important like not just so that a single agent can get better the next time you do it to do the exact same thing, but rather think about systems where you have hundreds or or thousands of agents that have some shared grouping of of tools that they have access to, right? So we’re able to persist these agent runs, these these

05:36 decision traces in the graph and then share that with other agents in our organization, right? so that we have one shared context graph uh that enables our agents to essentially learn from uh each other in this shared memory piece. Okay. So that that that’s the memory piece. Like this this is um uh a system that we’ve built at at Near Forj, a pattern that that lots of folks are are following for working uh with agent memory. But what about this idea of of

06:07 actionable knowledge, right? How do we enable our agents to to take action? Uh memory. One way to think of memory is is this is a good representation of what happened, right? We know um the people we were talking about, how they’re connected. we have the these decision traces. Um the next step is typically creating skills, right? How many people are are using skills with their agents today? Mo most folks. Cool. How many people have have written skills

06:38 yourself? Cool. Mo most folks. How many folks have had an agent write the skill for you? Right. Cool. Yeah. So that that’s what we’re we’re talking about here is going from this memory graph, this context graph to how can we use that to create like grounded uh executable skills. So if you’re not familiar with uh with a skill, this is um an open standard anthropic um put this out uh agentskills.io IO I think is where

07:11 the um open standard is hosted. And the basic idea here is that we have some uh some some metadata, right? Some description of like what this skill is about. Uh and then this progressive disclosure, right? So I I have lots of more detailed information. These are often in like markdown files, references that we can um progressively choose. If if we’re going down this path of one piece of the skill, we can uh retrieve and load that data uh into context. Right? So this is this then gives us some like sharable unit uh that we can

07:45 then uh take the skill, package it up uh and reuse that across different agents. But skills have a a similar challenge to some of the issues that we saw uh previously when when we were talking about uh memory, right? is that we’re we’re typically working with pros. Um, and so we’re still often limited in in this uh challenge of understanding is this the canonical thing that we’re talking about, making sure that we have like debugable steps and and output. Um

08:19 and so to address some of these challenges that uh we’re seeing with skills uh some folks on the NeoRaj research team um have done some interesting research and published some work on APE uh graph representation for learning and governing agent skills. Um this is a screenshot from the the paper. Zach is uh is here somewhere. If he’s not here, he’s he’s at um the NearJ booth today. So definitely uh chat with Zach if you’re interested in this. But

08:49 you can think of this uh APE is essentially an extension of the agent skills um protocol with some additional metadata that now treats your skills as typed execution graphs. Right? So we have uh steps modeled as nodes. We have uh very um very strict schema that governs how we describe that how these uh steps are actionable. Right? So think of think of this as a way of of

09:20 representing um a skill as a graph broken up into steps that are executable with a schema governed description. Um that’s the the ape protocol and this is from the uh the benchmark that was used in the the paper found that yeah like this actually matters when when they applied this to um skills bench saw like a significant increase in uh task successfully completed when looking at applying uh the a protocol to human uh curated uh

09:56 skills. So that looks interesting. How can we leverage some of those uh some of those ideas, some of that research for skill distillation? Uh so essentially what what we want to do is take this this memory graph, right? This context graph that’s maybe scoped to a a workspace or scoped to a a project in an organization and we want to distill that into a skill, but not just like a markdown

10:27 file. We want this to be grounded in actual data that we’ve observed. We want this to be like deterministic, right? We want uh well understood procedural steps in our graph. And we want to be able to govern this over time, right? If if the underlying data that makes up our skill changes in memory, we want to be able to understand that and and know about that. Um grounding is is an important piece here, right? making sure that the the data that makes up our skill is grounded

10:58 in our memory system. But that that’s that’s not enough for it to be useful, right? There are other heruristics that we need to look at here. So, uh things like coverage, are we making sure that the uh steps and descriptions of our um throughout our skill are grounded across the skill? Coherence. Coherence is interesting. This is a way that we can detect uh maybe if the um information the piece the subgraph going into the skill distillation is split across

11:30 multiple topics and we can suggest well you may want to create multiple skills here um and so on. So those are are some of the pieces that lead up to generating the skill. Uh skill governance is an important piece that I mentioned. Making sure that we can understand uh as that skill changes uh as the data changes are are we able to update the skill? And with dynamic loading, you can think of uh having sort of like a governed skills registry that allows us to retrieve uh

12:02 for any agent the the most recent up-to-date skill, understand uh if that skill may have been uh stale or or or drifted, right? Cool. So, we’ve implemented this in uh the NEFJ agent memory service or we call NAMS. This is um an agent memory as a service as part of our NearJ labs effort. Um we also have open source tooling around that that implements these patterns. And essentially the way this works is we we have a a workspace

12:36 scoped context graph, right? That has our three types of agent memory, short-term, long-term reasoning memory. Um we have uh background workers that are capable of going through this distillation process. We’ll take a look at at what this looks like in a minute. Um, and then having some background curation process that is still making sure to take care of that governance to surface when uh our skills become stale based on the data that we have in the

13:06 memory system. Um, this is the the pipeline that that we go through to generate these. Um, the I won’t go through this in in detail. The the one thing that we’re trying to point out here is that most of these steps are deterministic. Um we’re really only leveraging an LLM here uh for synthesizing some of the claims uh generating some of the the text that we use for some of the skill description. Uh and then the other thing I want to call out here is that the first piece is really identifying the scope like where

13:38 what is your starting scope for distilling one of these skills. Is it around a certain entity? the subgraph around that is it a specific conversation uh and that sort of thing. And so we can decompose uh the the way we represent the skill and make sure that each of these components is grounded again in underlying data. And that’s the piece that we’re looking at for if that skill uh essentially becomes stale. If any of that underlying data changes, is removed or uh becomes uh contradictory.

14:16 Um, cool. And then we can also, as I mentioned before, we can also compose skills, right? So, this is where that coherence piece comes in. We use um graph algorithms like community detection, right? So, if we’re mixing uh topics from multiple communities, that can be an indication that we need to decompose our skill and we’ll catch that in um the skill governance piece. Cool. So, I’ve got a few minutes left here. Let’s see what this looks like. Um so this is NAMS the near agent memory service. Um this is this is uh free

14:48 currently. Anyone can can sign in and and try this out. Um this is what the the dashboard looks like. The the basic idea here is that we have um REST API and MCP tools that we can expose to our agents that map to um ingesting, retrieving, working with long-term, short-term reasoning memory. Um we’re going through this entity extraction and resolution process. So I can uh look at

15:18 some graph representation of my uh agent memory here. I I I can traverse that and and so on. Um we can look at entities that have been flagged. Um one important piece here is this idea of an ontology. So here we’re using a healthcare ontology. So we we’re going to be working with data about you know encounters, providers, uh that sort of thing. Uh and then I have ingested a

15:49 bunch of conversations here. So we can see the conversations that have been um ingested related to a healthcare agent, right? And so now we’re ready to distill a skill. Um and we said the the first piece is to decide like what is the the scope of that of that skill. We can do this for you know an entire workspace. That’s often not what the case we want. We can do this around um you know a certain entity specific conversations a a class in the ontology. Let’s do this

16:22 around um our most recent conversation. And we’re going to see this is going to kick off um in the the distillation queue and go through and fetch the data, go through that sevenstage pipeline and um construct the skill as a graph and then package that up for us. Um here’s one. While this is running, let’s just take a look at this guy. So, here’s one uh that we

16:52 ran previously. This is uh was run on a patient intake uh conversation. And you can see here that we’ve essentially extracted out the steps that make up the skill to run from a patient intake um all the way through uh charting for the patient. But we can see each one of these steps is grounded in the actual tool calls and the entities that constructed the underlying components of the skill. And we packaged this up with skill um MD file. Uh if we downloaded

17:29 this, we would this would be packaged up with uh other references and and and so on. uh following that progressive disclosure uh standard that we use with agent skills. Cool. So that was a a quick look at kind of how we think about agent memory as part of this context graph uh with nearj and I’ll leave up some uh resources. You can c grab the slides here. There’s a link uh to the slides and a QR code. Um, the nearf agent memory service that I

18:05 mentioned um is listed here as well as lots of documentation uh and resources for some of our open source tooling. So that’s it. I’m out of time, but we have a NearJ booth uh so I will be there as well as lots of other folks from the NearJ team. So we’ll see you there. Thanks folks.