docs: update sop

Author: gusye1234

docs/docs.json

@@ -50,7 +50,8 @@
             "pages": [
               "learn/skill-space",
               "learn/search-skills",
-              "learn/tool"
+              "learn/tool",
+              "learn/design-complex"
             ]
           },
           {

docs/learn/design-complex.mdx

@@ -0,0 +1,31 @@
+---
+title: "Design a Complex Task Judgment"
+description: "Acontext allows you to add rules to judge what is a complex task."
+---
+
+Acontext won't just learn every success/failed tasks into the skill space. It only learns the complex tasks.
+
+## What is a Complex Task?
+
+By default, the task complexity is judged based on these conditions:
+- **(c.1)** Wrong tool parameter passing that could be avoided: **normal**
+- **(c.2)** Back-and-forth retries due to wrong strategy (not errors): **normal**
+- **(c.3)** User provides feedback/preferences to correct agent's wrong decision: **critical**
+- **(c.4)** User explicitly emphasized saving this workflow or experience: **critical**
+
+<Info>
+If one `critical` rule is satisfied, the task will be considered as a complex task.
+</Info>
+<Info>
+If at least two `normal` rules are satisfied, the task will be considered as a complex task.
+</Info>
+
+## Design your own Complexity Scoring Rules
+You can append more rules to enrich the complexity judgment, by passing `config.yaml`:
+```yaml config.yaml
+sop_agent_custom_scoring_rules:
+  - description: "User wants another theme of landing page"
+    level: "normal"
+  - description: "When the word 'save this as sop' appears in the user's message, it's a complex task"
+    level: "critical"
+```

docs/learn/tool.mdx

@@ -1,5 +1,5 @@
 ---
-title: "Tools in Space"
+title: "Edit Agent Tools"
 description: "Manage and organize tools that agents have learned and used across your projects"
 ---
 

docs/local.mdx

@@ -16,8 +16,30 @@ Once it's done, you can the following endpoints enabled:
 - Acontext Base URL: http://localhost:8029/api/v1
 - Acontext Dashboard: http://localhost:3000/
 
-## Try some Examples
-### Interactive mode
-```bash
-acontenxt create
-```
+## Launch with Envs
+Place your `.env` file and run `acontext docker up` at the same directory:
+```bash 
+cat .env
+acontext docker up
+```
+
+
+## Next Steps
+
+<CardGroup cols={2}>
+
+<Card title="Dashboard" icon="robot" href="/observe/dashboard">
+View your agent's tasks, conversations, and learned skills in one place
+</Card>
+<Card title="Store Messages" icon="message" href="/store/messages/multi-provider">
+Save agent conversations with context
+</Card>
+
+<Card title="Enable Learning" icon="sparkles" href="/learn/skill-space">
+Let your agent learn from experience
+</Card>
+
+<Card title="Configuration" icon="gear" href="/settings/runtime">
+Configure Acontext to your needs
+</Card>
+</CardGroup>

docs/observe/dashboard.mdx

@@ -80,7 +80,7 @@ The skill viewer helps you understand which capabilities your agent have self-le
   Learn how to obtain agent tasks in detail, from SDK.
 </Card>
 
-<Card title="Settings" icon="gear" href="/settings/core">
-  Customize your Acontext Server.
+<Card title="Self-learning" icon="sparkles" href="/learn/skill-space">
+  Learn how to enable your agent to self-learn skills and SOPs.
 </Card>
 </CardGroup>

src/server/core/tests/llm/test_sop_customization.py

@@ -3,7 +3,7 @@
 
 Tests custom scoring rules functionality.
 """
-import pytest
+
 from acontext_core.llm.prompt.sop_customization import SOPPromptCustomization
 from acontext_core.llm.prompt.task_sop import TaskSOPPrompt
 from acontext_core.schema.config import CustomScoringRule, ProjectConfig
@@ -21,16 +21,14 @@ def test_build_custom_scoring_section_empty(self):
     def test_build_custom_scoring_section_normal(self):
         """Test building custom scoring section with normal level rules"""
         rule1 = CustomScoringRule(
-            description="If the task involves database operations",
-            level="normal"
+            description="If the task involves database operations", level="normal"
         )
         rule2 = CustomScoringRule(
-            description="If the task requires file I/O",
-            level="normal"
+            description="If the task requires file I/O", level="normal"
         )
         customization = SOPPromptCustomization(custom_scoring_rules=[rule1, rule2])
         result = customization.build_custom_scoring_section(start_index=5)
-        
+
         assert "(c.5)" in result
         assert "(c.6)" in result
         assert "If the task involves database operations" in result
@@ -40,12 +38,11 @@ def test_build_custom_scoring_section_normal(self):
     def test_build_custom_scoring_section_critical(self):
         """Test building custom scoring section with critical level rules"""
         rule = CustomScoringRule(
-            description="If the task requires external API calls",
-            level="critical"
+            description="If the task requires external API calls", level="critical"
         )
         customization = SOPPromptCustomization(custom_scoring_rules=[rule])
         result = customization.build_custom_scoring_section(start_index=5)
-        
+
         assert "(c.5)" in result
         assert "If the task requires external API calls" in result
         assert "+ 3 points" in result  # critical level = 3 points
@@ -56,7 +53,7 @@ def test_build_custom_scoring_section_mixed(self):
         rule2 = CustomScoringRule(description="Critical rule", level="critical")
         customization = SOPPromptCustomization(custom_scoring_rules=[rule1, rule2])
         result = customization.build_custom_scoring_section(start_index=5)
-        
+
         assert "(c.5)" in result
         assert "(c.6)" in result
         assert "Normal rule" in result
@@ -76,7 +73,7 @@ def test_get_all_rule_indices_with_custom(self):
         rule2 = CustomScoringRule(description="Rule 2", level="critical")
         customization = SOPPromptCustomization(custom_scoring_rules=[rule1, rule2])
         indices = customization.get_all_rule_indices(base_count=4)
-        
+
         assert len(indices) == 6
         assert indices == ["(c.1)", "(c.2)", "(c.3)", "(c.4)", "(c.5)", "(c.6)"]
 
@@ -87,46 +84,44 @@ class TestTaskSOPPromptWithCustomization:
     def test_system_prompt_without_customization(self):
         """Test system prompt generation without customization"""
         prompt = TaskSOPPrompt.system_prompt()
-        
+
         # Should contain base rules
         assert "(c.1)" in prompt
         assert "(c.2)" in prompt
         assert "(c.3)" in prompt
         assert "(c.4)" in prompt
-        
+
         # Should not contain custom rules
         assert "(c.5)" not in prompt
-        
+
         # Report section should reference base rules only
         assert "Give your judgement on (c.1), (c.2), (c.3), (c.4)" in prompt
 
     def test_system_prompt_with_customization(self):
         """Test system prompt generation with customization"""
         rule1 = CustomScoringRule(
-            description="If the task involves database operations",
-            level="normal"
+            description="If the task involves database operations", level="normal"
         )
         rule2 = CustomScoringRule(
-            description="If the task requires external API calls",
-            level="critical"
+            description="If the task requires external API calls", level="critical"
         )
         customization = SOPPromptCustomization(custom_scoring_rules=[rule1, rule2])
         prompt = TaskSOPPrompt.system_prompt(customization=customization)
-        
+
         # Should contain base rules
         assert "(c.1)" in prompt
         assert "(c.2)" in prompt
         assert "(c.3)" in prompt
         assert "(c.4)" in prompt
-        
+
         # Should contain custom rules
         assert "(c.5)" in prompt
         assert "(c.6)" in prompt
         assert "If the task involves database operations" in prompt
         assert "If the task requires external API calls" in prompt
         assert "+ 1 point" in prompt  # normal
         assert "+ 3 points" in prompt  # critical
-        
+
         # Report section should reference all rules
         assert "(c.5)" in prompt
         assert "(c.6)" in prompt
@@ -138,13 +133,13 @@ def test_system_prompt_customization_appended_not_replaced(self):
         rule = CustomScoringRule(description="Custom rule", level="normal")
         customization = SOPPromptCustomization(custom_scoring_rules=[rule])
         prompt = TaskSOPPrompt.system_prompt(customization=customization)
-        
+
         # Base rules should still be present
         assert "(c.1)" in prompt
         assert "(c.2)" in prompt
         assert "(c.3)" in prompt
         assert "(c.4)" in prompt
-        
+
         # Custom rule should be appended
         assert "(c.5)" in prompt
         assert "Custom rule" in prompt
@@ -159,10 +154,8 @@ def test_project_config_with_custom_rules(self):
             CustomScoringRule(description="Test rule 1", level="normal"),
             CustomScoringRule(description="Test rule 2", level="critical"),
         ]
-        config = ProjectConfig(
-            sop_agent_custom_scoring_rules=custom_rules
-        )
-        
+        config = ProjectConfig(sop_agent_custom_scoring_rules=custom_rules)
+
         assert len(config.sop_agent_custom_scoring_rules) == 2
         assert config.sop_agent_custom_scoring_rules[0].description == "Test rule 1"
         assert config.sop_agent_custom_scoring_rules[0].level == "normal"
@@ -172,7 +165,6 @@ def test_project_config_with_custom_rules(self):
     def test_project_config_without_custom_rules(self):
         """Test ProjectConfig without custom scoring rules (default)"""
         config = ProjectConfig()
-        
+
         assert config.sop_agent_custom_scoring_rules == []
         assert isinstance(config.sop_agent_custom_scoring_rules, list)
-