@colbymchenry/cmem 0.5.1 → 0.5.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +44 -8
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -25,6 +25,22 @@ npx @colbymchenry/cmem
|
|
|
25
25
|
|
|
26
26
|
---
|
|
27
27
|
|
|
28
|
+
## Stop Writing Markdown Files
|
|
29
|
+
|
|
30
|
+
Everyone tries to solve Claude's memory problem the same way: **manually writing markdown files**. `CLAUDE.md`, `ARCHITECTURE.md`, decision logs, convention docs...
|
|
31
|
+
|
|
32
|
+
It's tedious. It's incomplete. And you forget to update them.
|
|
33
|
+
|
|
34
|
+
**cmem takes a different approach.** Instead of *you* documenting everything, cmem watches your conversations and *automatically* extracts the knowledge Claude needs. Architecture decisions, bug patterns, project conventions, anti-patterns—all captured naturally as you work.
|
|
35
|
+
|
|
36
|
+
Think of it like giving Claude a **notebook for each project**. Every conversation adds notes. When Claude needs to answer a question, it semantically searches this organized, efficient knowledge base—finding exactly the right context without you lifting a finger.
|
|
37
|
+
|
|
38
|
+
> **Traditional approach:** You manually write docs → Claude reads them → Docs aren't detailed enough so Claude explores anyway → You forget to update → Outdated docs lead to bad suggestions
|
|
39
|
+
>
|
|
40
|
+
> **cmem approach:** You just code → cmem learns from every session → Knowledge stays current and detailed → Claude gives informed answers instantly
|
|
41
|
+
|
|
42
|
+
---
|
|
43
|
+
|
|
28
44
|
## What is cmem?
|
|
29
45
|
|
|
30
46
|
**cmem** is a self-learning memory system for Claude Code. It:
|
|
@@ -33,15 +49,35 @@ npx @colbymchenry/cmem
|
|
|
33
49
|
2. **Learns from your sessions** — Auto-extracts lessons about your projects
|
|
34
50
|
3. **Injects relevant knowledge** — Claude starts each conversation with context about your codebase
|
|
35
51
|
4. **Searches your history** — Find past conversations by meaning, not keywords
|
|
36
|
-
5. **
|
|
52
|
+
5. **Rename and favorite sessions** - Never lose those valuable sessions between you and Claude
|
|
53
|
+
6. **View all sessions by project** - Organized chat sessions by what project you were working on
|
|
54
|
+
7. **Web GUI** — Manage learned knowledge through a beautiful interface
|
|
37
55
|
|
|
38
|
-
|
|
39
|
-
|
|
40
|
-
|
|
41
|
-
|
|
42
|
-
|
|
43
|
-
|
|
44
|
-
|
|
56
|
+
---
|
|
57
|
+
|
|
58
|
+
### MCP Integration
|
|
59
|
+
|
|
60
|
+
> **Using CMEM MCP to fetch conversational data about previous chat sessions**
|
|
61
|
+
>
|
|
62
|
+
> Claude can search through your conversation history and synthesize answers—all without cluttering your main session context.
|
|
63
|
+
|
|
64
|
+
|
|
65
|
+
|
|
66
|
+
---
|
|
67
|
+
|
|
68
|
+
### Session Navigation
|
|
69
|
+
|
|
70
|
+
> **Easily navigate between previous chat sessions and pick back up right where you left off**
|
|
71
|
+
>
|
|
72
|
+
> Browse, search, filter, rename, and restore any conversation with the interactive TUI.
|
|
73
|
+
|
|
74
|
+
|
|
75
|
+
|
|
76
|
+
#### Open all previous chat sessions in current directory
|
|
77
|
+
|
|
78
|
+
|
|
79
|
+
|
|
80
|
+
---
|
|
45
81
|
|
|
46
82
|
## Web GUI
|
|
47
83
|
|