@minns/openclaw-minns 0.2.1 → 0.3.2
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/dist/index.js +1 -1
- package/dist/index.js.map +1 -1
- package/openclaw.plugin.json +1 -1
- package/package.json +4 -3
- package/skills/minns-memory/SKILL.md +37 -0
package/dist/index.js
CHANGED
package/dist/index.js.map
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"index.js","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,eAAe;IACb,EAAE,EAAE,
|
|
1
|
+
{"version":3,"file":"index.js","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,eAAe;IACb,EAAE,EAAE,gBAAgB;IACpB,QAAQ,CAAC,GAAQ;QACf,iDAAiD;QACjD,6CAA6C;QAC7C,yCAAyC;IAC3C,CAAC;CACF,CAAC"}
|
package/openclaw.plugin.json
CHANGED
package/package.json
CHANGED
|
@@ -1,15 +1,16 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@minns/openclaw-minns",
|
|
3
|
-
"version": "0.2
|
|
3
|
+
"version": "0.3.2",
|
|
4
4
|
"description": "MCP stdio server integrating Minns EventGraphDB into OpenClaw",
|
|
5
5
|
"type": "module",
|
|
6
|
-
"main": "./dist/
|
|
7
|
-
"types": "./dist/
|
|
6
|
+
"main": "./dist/mcp.js",
|
|
7
|
+
"types": "./dist/mcp.d.ts",
|
|
8
8
|
"bin": {
|
|
9
9
|
"openclaw-minns-mcp": "./dist/mcp.js"
|
|
10
10
|
},
|
|
11
11
|
"files": [
|
|
12
12
|
"dist",
|
|
13
|
+
"skills",
|
|
13
14
|
"README.md",
|
|
14
15
|
"openclaw.plugin.json"
|
|
15
16
|
],
|
|
@@ -0,0 +1,37 @@
|
|
|
1
|
+
# Minns Memory Skill
|
|
2
|
+
|
|
3
|
+
Fallback policy skill for agents that do not yet have native MCP tool support.
|
|
4
|
+
When active, the agent should follow these rules automatically.
|
|
5
|
+
|
|
6
|
+
## Before answering
|
|
7
|
+
|
|
8
|
+
1. Call `memory.search` with a short query derived from the user's message.
|
|
9
|
+
2. If relevant memories are returned, incorporate them into your response.
|
|
10
|
+
3. If no memories match, proceed without referencing memory.
|
|
11
|
+
|
|
12
|
+
## After answering
|
|
13
|
+
|
|
14
|
+
1. Decide whether the interaction produced a **durable fact or preference** worth keeping.
|
|
15
|
+
- Examples: user timezone, preferred language, project conventions, confirmed decisions.
|
|
16
|
+
2. If yes, call `memory.capture` with a concise summary of the fact.
|
|
17
|
+
3. If the interaction was ephemeral (greetings, small talk, transient debug output), do **not** capture it.
|
|
18
|
+
|
|
19
|
+
## Safety rules
|
|
20
|
+
|
|
21
|
+
- **Never** store secrets, passwords, API keys, tokens, or credentials.
|
|
22
|
+
- **Never** store personally identifiable information (PII) unless the user explicitly asks you to remember it.
|
|
23
|
+
- Only store facts that are likely to remain true across sessions.
|
|
24
|
+
- Prefer short, precise text over long paragraphs when capturing memories.
|
|
25
|
+
|
|
26
|
+
## Tool reference
|
|
27
|
+
|
|
28
|
+
| Tool | When to use |
|
|
29
|
+
| ---------------------------- | ------------------------------------------------ |
|
|
30
|
+
| `memory.search` | Before answering -- recall relevant context |
|
|
31
|
+
| `memory.capture` | After answering -- persist durable facts |
|
|
32
|
+
| `memory.memories` | When the user asks "what do you know about me?" |
|
|
33
|
+
| `memory.strategies` | When reviewing learned strategies for an agent |
|
|
34
|
+
| `strategy.similar` | When looking for reusable approaches |
|
|
35
|
+
| `strategy.suggest_next_action` | When deciding what to do next in a workflow |
|
|
36
|
+
| `claims.search` | When verifying factual claims from past events |
|
|
37
|
+
| `health.check` | When diagnosing connectivity to EventGraphDB |
|