@mhrj/mcp-agent-loop 1.0.2 → 1.0.3
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 +11 -13
- package/package.json +2 -2
package/README.md
CHANGED
|
@@ -1,19 +1,21 @@
|
|
|
1
1
|
# Agentic Loop Memory Server ♾️
|
|
2
2
|
|
|
3
|
-
|
|
3
|
+
**The industry-standard persistent memory and state manager for long-running agentic workflows.**
|
|
4
4
|
|
|
5
|
-
|
|
5
|
+
Enable any AI model—especially smaller ones with limited context windows—to function with the persistence of high-end models. This project works as a two-part ecosystem: an **MCP Server** for state management and an **Agent Skill** for orchestration.
|
|
6
6
|
|
|
7
|
-
|
|
7
|
+
## 🛠 Complete Setup (Required)
|
|
8
8
|
|
|
9
|
+
For the best experience, you must install **both** the orchestration skill and the MCP server.
|
|
10
|
+
|
|
11
|
+
### 1. Install the Skill
|
|
12
|
+
Install instructions into your AI agent (Claude Code, Cursor, etc.):
|
|
9
13
|
```bash
|
|
10
14
|
npx skills add meharajM/agent-loop-mcp --yes
|
|
11
15
|
```
|
|
12
16
|
|
|
13
|
-
|
|
14
|
-
|
|
15
|
-
The skill requires the MCP server to be running. Add this to your `mcp_config.json`:
|
|
16
|
-
|
|
17
|
+
### 2. Configure the MCP Server
|
|
18
|
+
Add the following to your \`mcp_config.json\`:
|
|
17
19
|
```json
|
|
18
20
|
{
|
|
19
21
|
"mcpServers": {
|
|
@@ -25,12 +27,8 @@ The skill requires the MCP server to be running. Add this to your `mcp_config.js
|
|
|
25
27
|
}
|
|
26
28
|
```
|
|
27
29
|
|
|
28
|
-
## 🌟
|
|
29
|
-
|
|
30
|
-
- **Active Context Management**: Not just a search tool. It monitors your token/word count and forces a "Summarize & Compact" loop when needed.
|
|
31
|
-
- **Mandatory Self-Healing**: Every tool failure requires the AI to provide a new `self_heal_strategy`, preventing mindless retries.
|
|
32
|
-
- **Transparent Markdown Storage**: Your "brain" is stored in `~/.agent-loop-mcp/` as standard Markdown. You can inspect or edit it anytime.
|
|
33
|
-
- **Small Model Optimized**: Designed specifically for models like `Gemini Flash` and `GPT-4o-mini`.
|
|
30
|
+
## 🌟 Why this approach is unique
|
|
31
|
+
Unlike passive memory tools, this is an **Active State Manager**. It monitors word counts to trigger compaction cycles and enforces a "Self-Healing Strategy" on every failure, preventing AI agents from getting stuck in mindless loops.
|
|
34
32
|
|
|
35
33
|
## 📂 Project Structure
|
|
36
34
|
|
package/package.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@mhrj/mcp-agent-loop",
|
|
3
|
-
"version": "1.0.
|
|
3
|
+
"version": "1.0.3",
|
|
4
4
|
"mcpName": "io.github.meharajM/agent-loop-mcp",
|
|
5
|
-
"description": "Agentic memory
|
|
5
|
+
"description": "Agentic memory manager. Best used with skill: npx skills add meharajM/agent-loop-mcp",
|
|
6
6
|
"main": "build/index.js",
|
|
7
7
|
"bin": {
|
|
8
8
|
"mcp-agent-loop": "build/index.js"
|