AgentAlley

Skills · Coding · Agent

Knowledge Graph Memory Server

Unverified●29/40

Originally by modelcontextprotocol · MIT

Claude Code·UnknownFrontmatter could not be read
Cursor·UnknownWe have not crawled the repo tree, so we will not guess
Codex·UnknownWe have not crawled the repo tree, so we will not guess
Gemini CLI·UnknownThe spec defines no detection rule for Gemini
Copilot·UnknownWe have not crawled the repo tree, so we will not guess
npx agentalley add memory

This command does not work yet — the CLI is still being built. Until then, use Raw in the reader below to take the file.

Who is stuck, and on what

The whole source

No sign-in, no blur, nothing truncated
memory/SKILL.md352 lines10.6 KBRawView on GitHub
Frontmatter: no frontmatter — costs points on criterion B3.
1# Knowledge Graph Memory ServerA5 — No allowed-tools declared — no way to tell what this skill may touchB3 — Frontmatter: no frontmatter
2 
3A basic implementation of persistent memory using a local knowledge graph. This lets Claude remember information about the user across chats.
4 
5Published on npm as [`@modelcontextprotocol/server-memory`](https://www.npmjs.com/package/@modelcontextprotocol/server-memory).
6 
7## Core Concepts
8 
9### Entities
10Entities are the primary nodes in the knowledge graph. Each entity has:
11- A unique name (identifier)
12- An entity type (e.g., "person", "organization", "event")
13- A list of observations
14 
15Example:
16```json
17{
18 "name": "John_Smith",
19 "entityType": "person",
20 "observations": ["Speaks fluent Spanish"]
21}
22```
23 
24### Relations
25Relations define directed connections between entities. They are always stored in active voice and describe how entities interact or relate to each other.
26 
27Example:
28```json
29{
30 "from": "John_Smith",
31 "to": "Anthropic",
32 "relationType": "works_at"
33}
34```
35### Observations
36Observations are discrete pieces of information about an entity. They are:
37 
38- Stored as strings
39- Attached to specific entities
40- Can be added or removed independently
41- Should be atomic (one fact per observation)
42 
43Example:
44```json
45{
46 "entityName": "John_Smith",
47 "observations": [
48 "Speaks fluent Spanish",
49 "Graduated in 2019",
50 "Prefers morning meetings"
51 ]
52}
53```
54 
55## API
56 
57### Tools
58- **create_entities**
59 - Create multiple new entities in the knowledge graph
60 - Input: `entities` (array of objects)
61 - Each object contains:
62 - `name` (string): Entity identifier
63 - `entityType` (string): Type classification
64 - `observations` (string[]): Associated observations
65 - Ignores entities with existing names
66 
67- **create_relations**
68 - Create multiple new relations between entities
69 - Input: `relations` (array of objects)
70 - Each object contains:
71 - `from` (string): Source entity name
72 - `to` (string): Target entity name
73 - `relationType` (string): Relationship type in active voice
74 - Skips duplicate relations
75 - Fails if either the source or target entity doesn't exist
76 
77- **add_observations**
78 - Add new observations to existing entities
79 - Input: `observations` (array of objects)
80 - Each object contains:
81 - `entityName` (string): Target entity
82 - `contents` (string[]): New observations to add
83 - Returns added observations per entity
84 - Fails if entity doesn't exist
85 
86- **delete_entities**
87 - Remove entities and their relations
88 - Input: `entityNames` (string[])
89 - Cascading deletion of associated relations
90 - No error if an entity doesn't exist; the response reports which names were not found
91 
92- **delete_observations**
93 - Remove specific observations from entities
94 - Input: `deletions` (array of objects)
95 - Each object contains:
96 - `entityName` (string): Target entity
97 - `observations` (string[]): Observations to remove
98 - No error if an observation doesn't exist; the response reports how many were deleted
99 
100- **delete_relations**
101 - Remove specific relations from the graph
102 - Input: `relations` (array of objects)
103 - Each object contains:
104 - `from` (string): Source entity name
105 - `to` (string): Target entity name
106 - `relationType` (string): Relationship type
107 - No error if a relation doesn't exist; the response reports how many were deleted
108 
109- **read_graph**
110 - Read the entire knowledge graph
111 - No input required
112 - Returns complete graph structure with all entities and relations
113 
114- **search_nodes**
115 - Search for nodes based on query
116 - Input: `query` (string)
117 - Searches across:
118 - Entity names
119 - Entity types
120 - Observation content
121 - Returns matching entities and their relations
122 
123- **open_nodes**
124 - Retrieve specific nodes by name
125 - Input: `names` (string[])
126 - Returns:
127 - Requested entities
128 - Relations between requested entities
129 - Silently skips non-existent nodes
130 
131### Resources
132 
133- **knowledge-graph** (`memory://knowledge-graph`)
134 - The full knowledge graph as a readable MCP Resource
135 - MIME type: `application/json`
136 - Returns the same shape as `read_graph` (entities and relations)
137 - Mutation tools (`create_entities`, `create_relations`, `add_observations`, `delete_entities`, `delete_observations`, `delete_relations`) emit `notifications/resources/updated` for this URI, so subscribed clients see live changes
138 
139# Usage with Claude Desktop
140 
141### Setup
142 
143Add this to your claude_desktop_config.json:
144 
145#### Docker
146 
147```json
148{
149 "mcpServers": {
150 "memory": {
151 "command": "docker",
152 "args": ["run", "-i", "-v", "claude-memory:/app/dist", "--rm", "mcp/memory"]
153 }
154 }
155}
156```
157 
158#### NPX
159```json
160{
161 "mcpServers": {
162 "memory": {
163 "command": "npx",
164 "args": [
165 "-y",
166 "@modelcontextprotocol/server-memory"
167 ]
168 }
169 }
170}
171```
172 
173On Windows, use `cmd /c` to launch `npx`:
174 
175```json
176{
177 "mcpServers": {
178 "memory": {
179 "command": "cmd",
180 "args": [
181 "/c",
182 "npx",
183 "-y",
184 "@modelcontextprotocol/server-memory"
185 ]
186 }
187 }
188}
189```
190 
191#### NPX with custom setting
192 
193The server can be configured using the following environment variables:
194 
195```json
196{
197 "mcpServers": {
198 "memory": {
199 "command": "npx",
200 "args": [
201 "-y",
202 "@modelcontextprotocol/server-memory"
203 ],
204 "env": {
205 "MEMORY_FILE_PATH": "/path/to/custom/memory.jsonl"
206 }
207 }
208 }
209}
210```
211 
212On Windows, use:
213 
214```json
215{
216 "mcpServers": {
217 "memory": {
218 "command": "cmd",
219 "args": [
220 "/c",
221 "npx",
222 "-y",
223 "@modelcontextprotocol/server-memory"
224 ],
225 "env": {
226 "MEMORY_FILE_PATH": "/path/to/custom/memory.jsonl"
227 }
228 }
229 }
230}
231```
232 
233- `MEMORY_FILE_PATH`: Path to the memory storage JSONL file (default: `memory.jsonl` in the server directory)
234 
235# VS Code Installation Instructions
236 
237For quick installation, use one of the one-click installation buttons below:
238 
239[![Install with NPX in VS Code](https://img.shields.io/badge/VS_Code-NPM-0098FF?style=flat-square&logo=visualstudiocode&logoColor=white)](https://insiders.vscode.dev/redirect/mcp/install?name=memory&config=%7B%22command%22%3A%22npx%22%2C%22args%22%3A%5B%22-y%22%2C%22%40modelcontextprotocol%2Fserver-memory%22%5D%7D) [![Install with NPX in VS Code Insiders](https://img.shields.io/badge/VS_Code_Insiders-NPM-24bfa5?style=flat-square&logo=visualstudiocode&logoColor=white)](https://insiders.vscode.dev/redirect/mcp/install?name=memory&config=%7B%22command%22%3A%22npx%22%2C%22args%22%3A%5B%22-y%22%2C%22%40modelcontextprotocol%2Fserver-memory%22%5D%7D&quality=insiders)B1 — Line is 668 characters — unreadable by eye
240 
241[![Install with Docker in VS Code](https://img.shields.io/badge/VS_Code-Docker-0098FF?style=flat-square&logo=visualstudiocode&logoColor=white)](https://insiders.vscode.dev/redirect/mcp/install?name=memory&config=%7B%22command%22%3A%22docker%22%2C%22args%22%3A%5B%22run%22%2C%22-i%22%2C%22-v%22%2C%22claude-memory%3A%2Fapp%2Fdist%22%2C%22--rm%22%2C%22mcp%2Fmemory%22%5D%7D) [![Install with Docker in VS Code Insiders](https://img.shields.io/badge/VS_Code_Insiders-Docker-24bfa5?style=flat-square&logo=visualstudiocode&logoColor=white)](https://insiders.vscode.dev/redirect/mcp/install?name=memory&config=%7B%22command%22%3A%22docker%22%2C%22args%22%3A%5B%22run%22%2C%22-i%22%2C%22-v%22%2C%22claude-memory%3A%2Fapp%2Fdist%22%2C%22--rm%22%2C%22mcp%2Fmemory%22%5D%7D&quality=insiders)B1 — Line is 780 characters — unreadable by eye
242 
243For manual installation, you can configure the MCP server using one of these methods:
244 
245**Method 1: User Configuration (Recommended)**
246Add the configuration to your user-level MCP configuration file. Open the Command Palette (`Ctrl + Shift + P`) and run `MCP: Open User Configuration`. This will open your user `mcp.json` file where you can add the server configuration.
247 
248**Method 2: Workspace Configuration**
249Alternatively, you can add the configuration to a file called `.vscode/mcp.json` in your workspace. This will allow you to share the configuration with others.
250 
251> For more details about MCP configuration in VS Code, see the [official VS Code MCP documentation](https://code.visualstudio.com/docs/copilot/customization/mcp-servers).
252 
253#### NPX
254 
255```json
256{
257 "servers": {
258 "memory": {
259 "command": "npx",
260 "args": [
261 "-y",
262 "@modelcontextprotocol/server-memory"
263 ]
264 }
265 }
266}
267```
268 
269On Windows, use:
270 
271```json
272{
273 "servers": {
274 "memory": {
275 "command": "cmd",
276 "args": [
277 "/c",
278 "npx",
279 "-y",
280 "@modelcontextprotocol/server-memory"
281 ]
282 }
283 }
284}
285```
286 
287#### Docker
288 
289```json
290{
291 "servers": {
292 "memory": {
293 "command": "docker",
294 "args": [
295 "run",
296 "-i",
297 "-v",
298 "claude-memory:/app/dist",
299 "--rm",
300 "mcp/memory"
301 ]
302 }
303 }
304}
305```
306 
307### System Prompt
308 
309The prompt for utilizing memory depends on the use case. Changing the prompt will help the model determine the frequency and types of memories created.
310 
311Here is an example prompt for chat personalization. You could use this prompt in the "Custom Instructions" field of a [Claude.ai Project](https://www.anthropic.com/news/projects).
312 
313```
314Follow these steps for each interaction:
315 
3161. User Identification:
317 - You should assume that you are interacting with default_user
318 - If you have not identified default_user, proactively try to do so.
319 
3202. Memory Retrieval:
321 - Always begin your chat by saying only "Remembering..." and retrieve all relevant information from your knowledge graph
322 - Always refer to your knowledge graph as your "memory"
323 
3243. Memory
325 - While conversing with the user, be attentive to any new information that falls into these categories:
326 a) Basic Identity (age, gender, location, job title, education level, etc.)
327 b) Behaviors (interests, habits, etc.)
328 c) Preferences (communication style, preferred language, etc.)
329 d) Goals (goals, targets, aspirations, etc.)
330 e) Relationships (personal and professional relationships up to 3 degrees of separation)
331 
3324. Memory Update:
333 - If any new information was gathered during the interaction, update your memory as follows:
334 a) Create entities for recurring organizations, people, and significant events
335 b) Connect them to the current entities using relations
336 c) Store facts about them as observations
337```
338 
339## Building
340 
341Docker:
342 
343```sh
344docker build -t mcp/memory -f src/memory/Dockerfile .
345```
346 
347For Awareness: a prior mcp/memory volume contains an index.js file that could be overwritten by the new container. If you are using a docker volume for storage, delete the old docker volume's `index.js` file before starting the new container.
348 
349## License
350 
351This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License. For more details, please see the LICENSE file in the project repository.
352 

Reviews

Installed this one?Write the first review and take the Trailblazer badge.

Reviews only open after a real install, so this is empty — and we leave it empty rather than invent one.

Alternatives

Also in Coding
Subagent Driven DevelopmentUse when executing implementation plans with independent tasks in the current sessionCoding · MIT◐◐◐◐◐●36/40Python Code Style & DocumentationPython code style, linting, formatting, naming conventions, and documentation standards. Use when writing new code, reviewing style, configuring linters, writing docstrings, or establishing project standards.Coding · MIT◐····●35/40Competitor Price Analysis 💲Competitor pricing strategy analysis and market positioning. Price mapping, pricing gaps identification, elasticity signals evaluation, and strategic pricing optimization. Use when the user asks about competitor pricing, price analysis, pricing strategy, or coCoding · MIT◐····●34/40Competitor Price Tracker 📊Set up competitor price tracking and monitoring workflows. Track price changes, detect promotions, analyze pricing patterns, and get alerts for competitive price movements.Coding · MIT◐····●34/40