dialectic 0.1.0

package/examples/kata1/kata1.md

@@ -0,0 +1,5 @@
+A university has greatly expanded its CS course and wants to be able to automate the grading of simple programming assignments.
+
+Users: 300+ students per year, plus staff and admin.
+
+Requirements: Students must be able to upload their source code, which will be run and graded. Grades and runs must be persistent and auditable. There must be a plagiarism detection system involving comparing with other submissions and also submitting to a web-based service (TurnItIn). There must be some level of integration with the University's learning management system (LMS)

package/examples/kata1/meta.txt

@@ -0,0 +1 @@
+source: https://www.architecturalkatas.com/kata.html?kata=CheckYourWork.json

package/examples/kata2/debate-config.json

@@ -0,0 +1,54 @@
+{
+    "agents": [
+      {
+        "id": "agent-architect",
+        "name": "System Architect",
+        "role": "architect",
+        "model": "claude-sonnet-4",
+        "provider": "openai",
+        "temperature": 0.5,
+        "enabled": true
+      },
+      {
+        "id": "agent-performance",
+        "name": "Performance Engineer",
+        "role": "performance",
+        "model": "claude-sonnet-4",
+        "provider": "openai",
+        "temperature": 0.5,
+        "enabled": true
+      },
+      {
+        "id": "agent-security",
+        "name": "Security Specialist",
+        "role": "security",
+        "model": "claude-sonnet-4",
+        "provider": "openai",
+        "temperature": 0.5,
+        "enabled": true
+      }
+    ],
+    "judge": {
+      "id": "judge-main",
+      "name": "Technical Judge",
+      "role": "generalist",
+      "model": "claude-sonnet-4",
+      "provider": "openai",
+      "temperature": 0.5
+    },
+    "debate": {
+      "rounds": 3,
+      "terminationCondition": {
+        "type": "fixed"
+      },
+      "synthesisMethod": "judge",
+      "includeFullHistory": true,
+      "timeoutPerRound": 300000,
+      "summarization": {
+        "enabled": false,
+        "threshold": 2500,
+        "maxLength": 1250,
+        "method": "length-based"
+      }
+    }
+  }

package/examples/kata2/eval_config1.json

@@ -0,0 +1,21 @@
+{
+  "agents": [
+    {
+      "id": "eval-1",
+      "name": "Sonnet 4 Evaluator 1",
+      "model": "claude-sonnet-4",
+      "provider": "openai",
+      "timeout": 30000,
+      "enabled": true
+    },
+    {
+      "id": "eval-2",
+      "name": "Sonnet 4 Evaluator 2",
+      "model": "claude-sonnet-4",
+      "provider": "openai",
+      "timeout": 30000,
+      "enabled": true
+    }
+  ]
+}
+

package/examples/kata2/eval_config2.json

@@ -0,0 +1,25 @@
+{
+  "agents": [
+    {
+      "id": "eval-1",
+      "name": "Sonnet 4 Evaluator 1",
+      "model": "claude-sonnet-4",
+      "provider": "openai",
+      "timeout": 30000,
+      "enabled": true,
+      "systemPromptPath": "../eval_system.md",
+      "userPromptPath":"../eval_user.md"
+    },
+    {
+      "id": "eval-2",
+      "name": "Sonnet 4 Evaluator 2",
+      "model": "claude-sonnet-4",
+      "provider": "openai",
+      "timeout": 30000,
+      "enabled": true,
+      "systemPromptPath": "../eval_system.md",
+      "userPromptPath": "../eval_user.md"
+    }
+  ]
+}
+

package/examples/kata2/kata2.md

@@ -0,0 +1,5 @@
+Organization running "Mock UN" events wants to take its events online, permitting students to participate online
+
+Requirements: student-diplomats must be able to video-chat with one another; student-diplomats must be able to "give speeches" to the "assembly" (video-chat to the entire group); (mocked) world events (created by moderators) distributed via (mock) "news sites"; moderators must be able to monitor any video chat for appropriateness
+
+Users: 500 or so "diplomats" per "mock UN" gathering; dozens of moderators per "mock UN"; many "mock UN"s simultaneously; no new hardware requirements on students

package/examples/kata2/only_architect/debate-config.json

@@ -0,0 +1,45 @@
+{
+    "agents": [
+      {
+        "id": "agent-architect1",
+        "name": "System Architect 1",
+        "role": "architect",
+        "model": "claude-sonnet-4",
+        "provider": "openai",
+        "temperature": 0.5,
+        "enabled": true
+      },
+      {
+        "id": "agent-architect2",
+        "name": "System Architect 2",
+        "role": "architect",
+        "model": "claude-sonnet-4",
+        "provider": "openai",
+        "temperature": 0.5,
+        "enabled": true
+      }
+    ],
+    "judge": {
+      "id": "judge-main",
+      "name": "Technical Judge",
+      "role": "generalist",
+      "model": "claude-sonnet-4",
+      "provider": "openai",
+      "temperature": 0.5
+    },
+    "debate": {
+      "rounds": 3,
+      "terminationCondition": {
+        "type": "fixed"
+      },
+      "synthesisMethod": "judge",
+      "includeFullHistory": true,
+      "timeoutPerRound": 300000,
+      "summarization": {
+        "enabled": true,
+        "threshold": 10000,
+        "maxLength": 5000,
+        "method": "length-based"
+      }
+    }
+  }

package/examples/kata2/only_architect/eval_run.sh

@@ -0,0 +1,11 @@
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/only_architect/all_agents_1R_no_clarify.json -v -o examples/kata2/only_architect/eval_all_agents_1R_no_clarify.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/only_architect/all_agents_2R_no_clarify.json -v -o examples/kata2/only_architect/eval_all_agents_2R_no_clarify.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/only_architect/all_agents_3R_no_clarify.json -v -o examples/kata2/only_architect/eval_all_agents_3R_no_clarify.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/only_architect/all_agents_4R_no_clarify.json -v -o examples/kata2/only_architect/eval_all_agents_4R_no_clarify.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/only_architect/all_agents_5R_no_clarify.json -v -o examples/kata2/only_architect/eval_all_agents_5R_no_clarify.json -v
+
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/only_architect/all_agents_1R_no_clarify.json -v -o examples/kata2/only_architect/eval2_all_agents_1R_no_clarify.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/only_architect/all_agents_2R_no_clarify.json -v -o examples/kata2/only_architect/eval2_all_agents_2R_no_clarify.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/only_architect/all_agents_3R_no_clarify.json -v -o examples/kata2/only_architect/eval2_all_agents_3R_no_clarify.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/only_architect/all_agents_4R_no_clarify.json -v -o examples/kata2/only_architect/eval2_all_agents_4R_no_clarify.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/only_architect/all_agents_5R_no_clarify.json -v -o examples/kata2/only_architect/eval2_all_agents_5R_no_clarify.json -v

package/examples/kata2/only_architect/run_test.sh

@@ -0,0 +1,5 @@
+dialectic debate -r 1 -c examples/kata2/only_architect/debate-config.json -o examples/kata2/only_architect/all_agents_1R_no_clarify.json -p examples/kata2/kata2.md -v
+dialectic debate -r 2 -c examples/kata2/only_architect/debate-config.json -o examples/kata2/only_architect/all_agents_2R_no_clarify.json -p examples/kata2/kata2.md -v
+dialectic debate -r 3 -c examples/kata2/only_architect/debate-config.json -o examples/kata2/only_architect/all_agents_3R_no_clarify.json -p examples/kata2/kata2.md -v
+dialectic debate -r 4 -c examples/kata2/only_architect/debate-config.json -o examples/kata2/only_architect/all_agents_4R_no_clarify.json -p examples/kata2/kata2.md -v
+dialectic debate -r 5 -c examples/kata2/only_architect/debate-config.json -o examples/kata2/only_architect/all_agents_5R_no_clarify.json -p examples/kata2/kata2.md -v

package/examples/kata2/rounds_test/eval_run.sh

@@ -0,0 +1,11 @@
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/rounds_test/all_agents_1R_no_clarify.json -v -o examples/kata2/rounds_test/eval_all_agents_1R_no_clarify.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/rounds_test/all_agents_2R_no_clarify.json -v -o examples/kata2/rounds_test/eval_all_agents_2R_no_clarify.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/rounds_test/all_agents_3R_no_clarify.json -v -o examples/kata2/rounds_test/eval_all_agents_3R_no_clarify.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/rounds_test/all_agents_4R_no_clarify.json -v -o examples/kata2/rounds_test/eval_all_agents_4R_no_clarify.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/rounds_test/all_agents_5R_no_clarify.json -v -o examples/kata2/rounds_test/eval_all_agents_5R_no_clarify.json -v
+
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/rounds_test/all_agents_1R_no_clarify.json -v -o examples/kata2/rounds_test/eval2_all_agents_1R_no_clarify.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/rounds_test/all_agents_2R_no_clarify.json -v -o examples/kata2/rounds_test/eval2_all_agents_2R_no_clarify.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/rounds_test/all_agents_3R_no_clarify.json -v -o examples/kata2/rounds_test/eval2_all_agents_3R_no_clarify.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/rounds_test/all_agents_4R_no_clarify.json -v -o examples/kata2/rounds_test/eval2_all_agents_4R_no_clarify.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/rounds_test/all_agents_5R_no_clarify.json -v -o examples/kata2/rounds_test/eval2_all_agents_5R_no_clarify.json -v

package/examples/kata2/rounds_test/run_test.sh

@@ -0,0 +1,5 @@
+dialectic debate -r 1 -c examples/kata2/debate-config.json -o examples/kata2/rounds_test/all_agents_1R_no_clarify.json -p examples/kata2/kata2.md -v
+dialectic debate -r 2 -c examples/kata2/debate-config.json -o examples/kata2/rounds_test/all_agents_2R_no_clarify.json -p examples/kata2/kata2.md -v
+dialectic debate -r 3 -c examples/kata2/debate-config.json -o examples/kata2/rounds_test/all_agents_3R_no_clarify.json -p examples/kata2/kata2.md -v
+dialectic debate -r 4 -c examples/kata2/debate-config.json -o examples/kata2/rounds_test/all_agents_4R_no_clarify.json -p examples/kata2/kata2.md -v
+dialectic debate -r 5 -c examples/kata2/debate-config.json -o examples/kata2/rounds_test/all_agents_5R_no_clarify.json -p examples/kata2/kata2.md -v

package/examples/kata2/summary_length_test/eval_run.sh

@@ -0,0 +1,11 @@
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/summary_length_test/all_sum-length-1250.json -v -o examples/kata2/summary_length_test/eval_all_sum-length-1250.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/summary_length_test/all_sum-length-2500.json -v -o examples/kata2/summary_length_test/eval_all_sum-length-2500.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/summary_length_test/all_sum-length-5000.json -v -o examples/kata2/summary_length_test/eval_all_sum-length-5000.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/summary_length_test/all_sum-length-10000.json -v -o examples/kata2/summary_length_test/eval_all_sum-length-10000.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/summary_length_test/all_sum-length-15000.json -v -o examples/kata2/summary_length_test/eval_all_sum-length-15000.json -v
+
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/summary_length_test/all_sum-length-1250.json -v -o examples/kata2/summary_length_test/eval2_all_sum-length-1250.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/summary_length_test/all_sum-length-2500.json -v -o examples/kata2/summary_length_test/eval2_all_sum-length-2500.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/summary_length_test/all_sum-length-5000.json -v -o examples/kata2/summary_length_test/eval2_all_sum-length-5000.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/summary_length_test/all_sum-length-10000.json -v -o examples/kata2/summary_length_test/eval2_all_sum-length-10000.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/summary_length_test/all_sum-length-15000.json -v -o examples/kata2/summary_length_test/eval2_all_sum-length-15000.json -v

package/examples/kata2/summary_length_test/eval_run_w_clarify.sh

@@ -0,0 +1,7 @@
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/summary_length_test/all_sum-length-1250_clarify.json -v -o examples/kata2/summary_length_test/eval_all_sum-length-1250_clarify.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/summary_length_test/all_sum-length-5000_clarify.json -v -o examples/kata2/summary_length_test/eval_all_sum-length-5000_clarify.json -v
+# dialectic eval -c ./examples/kata2/eval_config1.json -d ./examples/kata2/summary_length_test/all_sum-length-10000_clarify.json -v -o examples/kata2/summary_length_test/eval_all_sum-length-10000_clarify.json -v
+
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/summary_length_test/all_sum-length-1250_clarify.json -v -o examples/kata2/summary_length_test/eval2_all_sum-length-1250_clarify.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/summary_length_test/all_sum-length-5000_clarify.json -v -o examples/kata2/summary_length_test/eval2_all_sum-length-5000_clarify.json -v
+dialectic eval -c ./examples/kata2/eval_config2.json -d ./examples/kata2/summary_length_test/all_sum-length-10000_clarify.json -v -o examples/kata2/summary_length_test/eval2_all_sum-length-10000_clarify.json -v

package/examples/kata2/summary_length_test/run_test.sh

@@ -0,0 +1,5 @@
+dialectic debate -r 3 -c examples/kata2/summary_length_test/debate-config-sum-t-1250.json -o examples/kata2/summary_length_test/all_sum-length-1250.json -p examples/kata2/kata2.md -v
+dialectic debate -r 3 -c examples/kata2/summary_length_test/debate-config-sum-t-2500.json -o examples/kata2/summary_length_test/all_sum-length-2500.json -p examples/kata2/kata2.md -v
+dialectic debate -r 3 -c examples/kata2/summary_length_test/debate-config-sum-t-5000.json -o examples/kata2/summary_length_test/all_sum-length-5000.json -p examples/kata2/kata2.md -v
+dialectic debate -r 3 -c examples/kata2/summary_length_test/debate-config-sum-t-10000.json -o examples/kata2/summary_length_test/all_sum-length-10000.json -p examples/kata2/kata2.md -v
+dialectic debate -r 3 -c examples/kata2/summary_length_test/debate-config-sum-t-15000.json -o examples/kata2/summary_length_test/all_sum-length-15000.json -p examples/kata2/kata2.md -v

package/examples/task-queue/debate-config.json

@@ -0,0 +1,76 @@
+{
+  "agents": [
+    {
+      "id": "agent-architect",
+      "name": "System Architect",
+      "role": "architect",
+      "model": "gpt-4",
+      "provider": "openai",
+      "temperature": 0.5,
+      "enabled": true,
+      "summarization": {
+        "enabled": true,
+        "threshold": 4000,
+        "maxLength": 2000,
+        "method": "length-based"
+      }
+    },
+    {
+      "id": "agent-performance",
+      "name": "Performance Engineer",
+      "role": "performance",
+      "model": "gpt-4",
+      "provider": "openai",
+      "temperature": 0.5,
+      "enabled": true,
+      "summarization": {
+        "enabled": true,
+        "threshold": 3500,
+        "maxLength": 1750,
+        "method": "length-based"
+      }
+    },
+    {
+      "id": "agent-security",
+      "name": "Security Specialist",
+      "role": "security",
+      "model": "gpt-4",
+      "provider": "openai",
+      "temperature": 0.4,
+      "enabled": true,
+      "summarization": {
+        "enabled": true,
+        "threshold": 3000,
+        "maxLength": 1500,
+        "method": "length-based"
+      }
+    }
+  ],
+  "judge": {
+    "id": "judge-main",
+    "name": "Technical Judge",
+    "role": "generalist",
+    "model": "gpt-4",
+    "provider": "openai",
+    "temperature": 0.3,
+    "summarization": {
+      "enabled": true,
+      "threshold": 6000,
+      "maxLength": 3000,
+      "method": "length-based"
+    }
+  },
+  "debate": {
+    "rounds": 4,
+    "terminationCondition": { "type": "fixed" },
+    "synthesisMethod": "judge",
+    "includeFullHistory": true,
+    "timeoutPerRound": 600000,
+    "summarization": {
+      "enabled": true,
+      "threshold": 5000,
+      "maxLength": 2500,
+      "method": "length-based"
+    }
+  }
+}