chore(agents): sync agent configs, models, capability index; cleanup junk

- Update agent model assignments (minimax/glm -> nemotron-3-ultra, kimi-k2.7-code, qwen3.5:397b) in .kilo/agents, kilo-meta.json, kilo.jsonc, capability-index.yaml
- Update orchestrator/agent prompts (complexity fast-path, verification tests, close-loop audit)
- Add .kilo/KILO_SPEC.md (Kilo Code specification reference)
- AGENTS.md: consolidate smartadmin agent rows
- Remove screenshot-dash.cjs (unused, contained hardcoded admin token); gitignore it
- Remove empty .kilo/milestones/

.kilo/capability-index.yaml

@@ -13,7 +13,7 @@ agents:
     produces: [code, documentation_inline]
 
   frontend-developer:
-    model: ollama-cloud/minimax-m2.5
+    model: ollama-cloud/qwen3.5:397b
     variant: thinking
     mode: subagent
     capabilities: [ui_implementation, component_creation, styling, responsive_design, nextjs_development, vue_nuxt_development, react_development]
@@ -47,7 +47,7 @@ agents:
     produces: [express_routes, database_schema, api_documentation]
 
   go-developer:
-    model: ollama-cloud/kimi-k2.6
+    model: ollama-cloud/kimi-k2.7-code
     mode: subagent
     capabilities: [go_api_development, go_database_design, go_concurrent_programming, go_authentication, go_microservices, postgresql_integration, sqlite_integration, clickhouse_integration]
     forbidden: [frontend_code]
@@ -55,7 +55,7 @@ agents:
     produces: [go_handlers, go_database_schema, go_api_documentation, concurrent_solutions]
 
   flutter-developer:
-    model: ollama-cloud/minimax-m2.5
+    model: ollama-cloud/qwen3.5:397b
     variant: thinking
     mode: subagent
     capabilities: [dart_programming, flutter_ui, mobile_app_development, widget_creation, state_management]
@@ -82,7 +82,7 @@ agents:
     produces: [test_files, test_reports, coverage_reports]
 
   code-skeptic:
-    model: ollama-cloud/glm-5.2
+    model: ollama-cloud/kimi-k2.7-code
     variant: thinking
     mode: subagent
     capabilities: [code_review, security_review, style_check, issue_identification]
@@ -91,7 +91,7 @@ agents:
     produces: [review_comments, approval_status, issue_list]
 
   security-auditor:
-    model: ollama-cloud/glm-5.2
+    model: ollama-cloud/kimi-k2.7-code
     variant: thinking
     mode: subagent
     capabilities: [vulnerability_scan, owasp_check, secret_detection, auth_review]
@@ -109,7 +109,7 @@ agents:
     produces: [performance_report, optimization_suggestions]
 
   the-fixer:
-    model: ollama-cloud/glm-5.2
+    model: ollama-cloud/kimi-k2.7-code
     variant: thinking
     mode: subagent
     capabilities: [bug_fixing, issue_resolution, code_correction]
@@ -118,7 +118,7 @@ agents:
     produces: [code_fixes, resolution_notes]
 
   browser-automation:
-    model: ollama-cloud/minimax-m3
+    model: ollama-cloud/kimi-k2.7-code
     variant: thinking
     mode: subagent
     capabilities: [e2e_browser_tests, form_filling, navigation_testing, screenshot_capture]
@@ -127,7 +127,7 @@ agents:
     produces: [test_results, screenshots]
 
   visual-tester:
-    model: ollama-cloud/kimi-k2.6
+    model: ollama-cloud/kimi-k2.7-code
     variant: thinking
     mode: subagent
     capabilities: [visual_regression, pixel_comparison, screenshot_diff, ui_validation, bbox_element_extraction, console_error_detection, network_error_detection, responsive_layout_check, button_overflow_detection, gitea_integration, docker_networking]
@@ -154,7 +154,7 @@ agents:
     produces: [analysis_report, recommendations, new_agent_specs]
 
   orchestrator:
-    model: ollama-cloud/glm-5.1
+    model: ollama-cloud/deepseek-v4-flash:0731
     variant: thinking
     mode: all
     capabilities: [task_routing, state_management, agent_coordination, workflow_execution]
@@ -163,7 +163,7 @@ agents:
     produces: [routing_decisions, status_updates]
 
   intake-agent:
-    model: ollama-cloud/minimax-m2.7
+    model: ollama-cloud/nemotron-3-ultra
     variant: thinking
     mode: all
     capabilities: [natural_language_understanding, conversational_clarification, task_structuring, requirement_formulation, intent_extraction]
@@ -172,7 +172,7 @@ agents:
     produces: [structured_tasks, acceptance_criteria, orchestrator_compatible_tasks]
 
   context-compressor:
-    model: ollama-cloud/minimax-m2.7
+    model: ollama-cloud/nemotron-3-ultra
     variant: thinking
     mode: subagent
     capabilities: [context_summarization, state_extraction, token_efficiency, memory_pruning, checkpoint_preservation]
@@ -181,7 +181,7 @@ agents:
     produces: [compressed_checkpoint, token_savings, preserved_state]
 
   pattern-matcher:
-    model: ollama-cloud/minimax-m2.7
+    model: ollama-cloud/nemotron-3-ultra
     variant: thinking
     mode: subagent
     capabilities: [similar_pattern_detection, successful_solution_retrieval, cross_project_learning, proactive_recommendation]
@@ -190,7 +190,7 @@ agents:
     produces: [pattern_matches, recommendations, knowledge_graph_updates]
 
   stakeholder-bridge:
-    model: ollama-cloud/minimax-m2.7
+    model: ollama-cloud/nemotron-3-ultra
     variant: thinking
     mode: subagent
     capabilities: [technical_to_business_translation, executive_summary_generation, progress_report_generation, stakeholder_communication]
@@ -199,7 +199,7 @@ agents:
     produces: [executive_summary, progress_report, business_translation]
 
   release-manager:
-    model: ollama-cloud/glm-5.2
+    model: ollama-cloud/deepseek-v4-flash:0731
     mode: subagent
     capabilities: [git_operations, version_management, changelog_creation, deployment]
     forbidden: [code_changes, feature_development]
@@ -225,7 +225,7 @@ agents:
     produces: [improved_prompts, optimization_report]
 
   product-owner:
-    model: ollama-cloud/minimax-m2.5
+    model: ollama-cloud/nemotron-3-ultra
     variant: thinking
     mode: subagent
     capabilities: [issue_management, prioritization, backlog_management, workflow_completion]
@@ -234,7 +234,7 @@ agents:
     produces: [priority_order, issue_labels, issue_closures]
 
   pipeline-judge:
-    model: ollama-cloud/qwen3.5:397b
+    model: ollama-cloud/kimi-k2.7-code
     variant: thinking
     mode: subagent
     capabilities: [test_execution, fitness_scoring, metric_collection, bottleneck_detection]
@@ -243,7 +243,7 @@ agents:
     produces: [fitness_report, bottleneck_analysis, improvement_triggers]
 
   workflow-architect:
-    model: ollama-cloud/glm-5.2
+    model: ollama-cloud/minimax-m3
     variant: thinking
     mode: subagent
     capabilities: [workflow_design, process_definition, automation_setup]
@@ -252,7 +252,7 @@ agents:
     produces: [workflow_definitions, command_files]
 
   markdown-validator:
-    model: ollama-cloud/minimax-m2.5
+    model: ollama-cloud/nemotron-3-ultra
     variant: thinking
     mode: subagent
     capabilities: [markdown_validation, formatting_check, link_validation]
@@ -279,7 +279,7 @@ agents:
     produces: [decomposed_steps, dependency_graph, success_criteria]
 
   reflector:
-    model: ollama-cloud/glm-5.2
+    model: ollama-cloud/minimax-m3
     variant: thinking
     mode: subagent
     capabilities: [self_reflection, mistake_analysis, lesson_extraction, trajectory_analysis, heuristic_evaluation]
@@ -296,7 +296,7 @@ agents:
     produces: [retrieved_memories, relevance_scores, consolidated_memories]
 
   architect-indexer:
-    model: ollama-cloud/glm-5.2
+    model: ollama-cloud/deepseek-v4-flash:0731
     mode: subagent
     capabilities: [codebase_indexing, project_mapping, architecture_documentation, dependency_analysis, entity_extraction, api_surface_discovery, convention_detection, staleness_detection]
     forbidden: [code_changes, implementation]
@@ -304,7 +304,7 @@ agents:
     produces: [.architect/state.json, .architect/project.json, .architect/README.md, architecture_overview, dependency_graph, entity_documentation, db_schema_documentation, api_surface_documentation, convention_documentation, file_graph, module_graph]
 
   history-miner:
-    model: ollama-cloud/glm-5.2
+    model: ollama-cloud/deepseek-v4-flash:0731
     mode: subagent
     capabilities: [git_history_analysis, duplicate_detection, regression_prevention, pattern_matching, past_solution_retrieval]
     forbidden: [implementation]
@@ -312,7 +312,7 @@ agents:
     produces: [historical_findings, regression_warnings, recommended_solutions]
 
   incident-responder:
-    model: ollama-cloud/glm-5.2
+    model: ollama-cloud/minimax-m3
     variant: thinking
     mode: subagent
     capabilities: [incident_response, live_forensics, malware_removal, persistence_hunting, ssh_cleanup, post_incident_hardening, cross_platform_hardening]
@@ -339,7 +339,7 @@ agents:
     produces: [evaluation_scores, detailed_commentary]
 
   smartadmin-builder:
-    model: ollama-cloud/minimax-m2.5
+    model: ollama-cloud/qwen3.5:397b
     variant: thinking
     mode: subagent
     capabilities: [smartadmin_template_building, ejs_generation, admin_panel_creation, component_library_usage]
@@ -348,7 +348,7 @@ agents:
     produces: [ejs_templates, admin_pages, smartadmin_components]
 
   smartadmin-viz-agent:
-    model: ollama-cloud/deepseek-v4-pro
+    model: ollama-cloud/deepseek-v4-flash:0731
     variant: thinking
     mode: subagent
     capabilities: [data_visualization, apexcharts, peity, easy_pie_chart, smarttable, kpi_cards]
@@ -357,7 +357,7 @@ agents:
     produces: [chart_html_snippets, chart_init_js, kpi_cards, smarttable_blocks]
 
   smartadmin-notify-agent:
-    model: ollama-cloud/glm-5.2
+    model: ollama-cloud/deepseek-v4-flash:0731
     variant: thinking
     mode: subagent
     capabilities: [notification_ui, alerts, toasts, modals, confirmations, inline_messages]
@@ -366,7 +366,7 @@ agents:
     produces: [alert_html, toast_html, modal_html, trigger_api, dismiss_handlers]
 
   smartadmin-form-agent:
-    model: ollama-cloud/minimax-m2.5
+    model: ollama-cloud/qwen3.5:397b
     variant: thinking
     mode: subagent
     capabilities: [form_generation, form_validation, select2, datepickers, form_wizards, field_dependencies]
@@ -375,7 +375,7 @@ agents:
     produces: [form_html, validation_js, dependency_handlers, init_code]
 
   smartadmin-interactive-agent:
-    model: ollama-cloud/kimi-k2.6
+    model: ollama-cloud/kimi-k2.7-code
     variant: thinking
     mode: subagent
     capabilities: [interactive_elements, buttons, dropdowns, nav_tabs, accordions, collapse, modal_triggers, state_toggles]
@@ -384,7 +384,7 @@ agents:
     produces: [element_html, event_handlers, state_api, widget_init]
 
   requirement-refiner:
-    model: ollama-cloud/glm-5.2
+    model: ollama-cloud/minimax-m3
     variant: thinking
     mode: all
     capabilities: [requirements_analysis, user_story_creation, acceptance_criteria_definition, technical_constraint_identification]
@@ -581,4 +581,53 @@ iteration_loops:
     evaluator: pipeline-judge
     optimizer: prompt-optimizer
     max_iterations: 3
-    convergence: fitness_above_0.85
+    convergence: fitness_above_0.85
+
+# Adaptive scaling by task complexity
+adaptive_scaling:
+  trivial:
+    consensus: false
+    reviewers: 1
+    max_iterations: 1
+    token_budget: 2000
+  simple:
+    consensus: false
+    reviewers: 1
+    max_iterations: 2
+    token_budget: 5000
+  medium:
+    consensus: false
+    reviewers: 2
+    reviewer_models: [ollama-cloud/glm-5.2, ollama-cloud/deepseek-v4-pro]
+    max_iterations: 3
+    token_budget: 10000
+  complex:
+    consensus: true
+    consensus_agents: 3
+    consensus_models: [ollama-cloud/glm-5.2, ollama-cloud/deepseek-v4-pro, ollama-cloud/minimax-m3]
+    reviewers: 2
+    max_iterations: 3
+    token_budget: 20000
+  complexity_routing:
+    source: requirement-refiner
+    field: complexity
+    values: [trivial, simple, medium, complex]
+
+# Consensus groups for critical decisions (Agent Forest pattern)
+consensus_groups:
+  architecture_decision:
+    agents: [system-analyst, planner, capability-analyst]
+    strategy: weighted_majority
+    weights: [0.4, 0.3, 0.3]
+    threshold: 0.65
+    trigger: status_researching
+  security_review:
+    agents: [security-auditor, security-auditor, security-auditor]
+    models: [ollama-cloud/glm-5.2, ollama-cloud/deepseek-v4-pro, ollama-cloud/minimax-m3]
+    strategy: majority_vote
+    trigger: security_review_phase
+  code_review:
+    agents: [code-skeptic, code-skeptic]
+    models: [ollama-cloud/glm-5.2, ollama-cloud/deepseek-v4-pro]
+    strategy: unanimous_required
+    trigger: status_implementing