UniqueSoft/APAW
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

evolution: objective model assignments from benchmark research + code-skeptic review

Browse Source 
- 6 model mismatches fixed: product-owner, incident-responder, history-miner,
  architect-indexer, pipeline-judge, workflow-cross-checker
- incident-responder capabilities corrected (was copy-pasted from workflow-cross-checker)
- history-miner entry added to capability-index.yaml
- LLM Stats API metadata ingested into research-report.json
- planner rationale corrected (removed false minimax-m3 "300-agent swarm" claim)
- capability-index.yaml: stale qwen3-coder:480b fallback_models removed (4 agents)
- capability-index.yaml: duplicate kimi-k2.6 entry removed (evolution-prompt)
- sync-agents.cjs --check: zero violations
This commit is contained in:
Deploy Bot
2026-06-01 22:13:49 +01:00
parent 397d8367e9
commit c1e50495a9
12 changed files with 182 additions and 71 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
.kilo/KILO_SPEC.md
View File

@@ -434,7 +434,7 @@ Provider availability depends on configuration. Common providers include:
| Agent | Role | Model |
|-------|------|-------|
| `@RequirementRefiner` | Converts vague ideas and bug reports into strict User Stories with acceptance criteria checklists. | ollama-cloud/deepseek-v4-pro |
| `@HistoryMiner` | Analyzes git history to find duplicates and past solutions, preventing regression and duplicate work. | ollama-cloud/qwen3-coder:480b |
| `@HistoryMiner` | Analyzes git history to find duplicates and past solutions, preventing regression and duplicate work. | ollama-cloud/deepseek-v4-pro |
| `@SystemAnalyst` | Designs technical specifications, data schemas, and API contracts before implementation. | ollama-cloud/minimax-m3:cloud |
| `@SdetEngineer` | Writes tests following TDD methodology. | ollama-cloud/deepseek-v4-pro |
| `@LeadDeveloper` | Primary code writer for backend and core logic. | ollama-cloud/deepseek-v4-pro |
@@ -451,7 +451,7 @@ Provider availability depends on configuration. Common providers include:
| `@ReleaseManager` | Manages git operations, semantic versioning, branching, and deployments. | ollama-cloud/deepseek-v4-pro |
| `@Evaluator` | Scores agent effectiveness after task completion for continuous improvement. | ollama-cloud/deepseek-v4-pro |
| `@PromptOptimizer` | Improves agent system prompts based on performance failures. | ollama-cloud/minimax-m3:cloud |
| `@ProductOwner` | Manages issue checklists, status labels, tracks progress and coordinates with human users. | ollama-cloud/kimi-k2.6 |
| `@ProductOwner` | Manages issue checklists, status labels, tracks progress and coordinates with human users. | ollama-cloud/minimax-m2.5:cloud |
| `@AgentArchitect` | Creates, modifies, and reviews new agents, workflows, and skills based on capability gap analysis. | ollama-cloud/minimax-m3:cloud |
| `@CapabilityAnalyst` | Analyzes task requirements against available agents, workflows, and skills. | ollama-cloud/minimax-m3:cloud |
| `@WorkflowArchitect` | Creates and maintains workflow definitions with complete architecture, Gitea integration, and quality gates. | ollama-cloud/glm-5.1 |
@@ -460,13 +460,13 @@ Provider availability depends on configuration. Common providers include:
| `@Planner` | Advanced task planner using Chain of Thought, Tree of Thoughts, and Plan-Execute-Reflect. | ollama-cloud/minimax-m3:cloud |
| `@Reflector` | Self-reflection agent using Reflexion pattern - learns from mistakes. | ollama-cloud/glm-5.1 |
| `@MemoryManager` | Manages agent memory systems - short-term (context), long-term (vector store), and episodic (experiences). | ollama-cloud/minimax-m3:cloud |
| `@ArchitectIndexer` | Indexes and maps project codebase architecture into . | ollama-cloud/qwen3-coder:480b |
| `@ArchitectIndexer` | Indexes and maps project codebase architecture into . | ollama-cloud/deepseek-v4-pro |
| `@FlutterDeveloper` | Flutter mobile specialist for cross-platform apps, state management, and UI components. | ollama-cloud/minimax-m2.5:cloud |
| `@PhpDeveloper` | PHP specialist for Laravel, Symfony, WordPress, and modular architecture. | ollama-cloud/deepseek-v4-pro |
| `@PipelineJudge` | Automated pipeline judge. | ollama-cloud/qwen3-coder:480b |
| `@PipelineJudge` | Automated pipeline judge. | ollama-cloud/deepseek-v4-pro |
| `@PythonDeveloper` | Python specialist for Django, FastAPI, data processing, and ML pipelines. | ollama-cloud/deepseek-v4-pro |
| `@IncidentResponder` | Server incident response and system hardening specialist. | ollama-cloud/deepseek-v4-pro |
| `@WorkflowCrossChecker` | Workflow cross-checker and process inspector. | ollama-cloud/qwen3-coder:480b |
| `@IncidentResponder` | Server incident response and system hardening specialist. | ollama-cloud/glm-5.1 |
| `@WorkflowCrossChecker` | Workflow cross-checker and process inspector. | ollama-cloud/deepseek-v4-pro |
| `@EvolutionSkeptic` | Evaluates model responses against role-specific rubrics with detailed scoring and commentary. | ollama-cloud/deepseek-v4-pro |
| `@EvolutionPrompt` | Generates role-specific stress-test prompts by analyzing agent definitions. | ollama-cloud/minimax-m3:cloud |

2
.kilo/agents/architect-indexer.md
View File

@@ -1,7 +1,7 @@
---
description: Indexes and maps project codebase architecture into .architect/ directory. Creates and maintains structured documentation of entities, APIs, DB schema, file graphs, and conventions. (GNS-2 Tier 0)
mode: subagent
model: ollama-cloud/qwen3-coder:480b
model: ollama-cloud/deepseek-v4-pro
variant: thinking
color: "#10B981"
permission:

2
.kilo/agents/history-miner.md
View File

@@ -1,7 +1,7 @@
---
description: Analyzes git history to find duplicates and past solutions, preventing regression and duplicate work (GNS-2 Tier 0)
mode: subagent
model: ollama-cloud/qwen3-coder:480b
model: ollama-cloud/deepseek-v4-pro
color: "#059669"
permission:
read: allow

2
.kilo/agents/incident-responder.md
View File

@@ -1,7 +1,7 @@
---
description: Server incident response and system hardening specialist. Handles live forensics, malware removal, persistence hunting, SSH-based server cleanup, and post-incident hardening. Works with any OS and panel.
mode: subagent
model: ollama-cloud/deepseek-v4-pro
model: ollama-cloud/glm-5.1
color: "#B91C1C"
permission:
read: allow

2
.kilo/agents/pipeline-judge.md
View File

@@ -1,7 +1,7 @@
---
description: Automated pipeline judge. Evaluates workflow execution by running tests, measuring token cost and wall-clock time. Produces objective fitness scores. Never writes code - only measures and scores. (GNS-2 Tier 0)
mode: subagent
model: ollama-cloud/qwen3-coder:480b
model: ollama-cloud/deepseek-v4-pro
color: "#DC2626"
permission:
read: allow

2
.kilo/agents/product-owner.md
View File

@@ -1,7 +1,7 @@
---
description: Manages issue checklists, status labels, tracks progress and coordinates with human users (GNS-2 Tier 1)
mode: subagent
model: ollama-cloud/kimi-k2.6
model: ollama-cloud/minimax-m2.5:cloud
color: "#EA580C"
permission:
read: allow

2
.kilo/agents/workflow-cross-checker.md
View File

@@ -1,7 +1,7 @@
---
description: Workflow cross-checker and process inspector. Analyzes inter-agent interaction logic, prevents conflicting tasks between agents, validates conformance to project architecture, tracks current state, and asks uncomfortable but important questions before expensive work begins.
mode: subagent
model: ollama-cloud/qwen3-coder:480b
model: ollama-cloud/deepseek-v4-pro
variant: thinking
color: "#9333EA"
permission:

71
.kilo/capability-index.yaml
View File

@@ -22,7 +22,6 @@ agents:
- code-skeptic
- orchestrator
fallback_models:
- ollama-cloud/qwen3-coder:480b
- ollama-cloud/kimi-k2.6
- ollama-cloud/deepseek-v4-pro
failover_strategy: downgraded
@@ -296,7 +295,6 @@ agents:
- performance-engineer
- orchestrator
fallback_models:
- ollama-cloud/qwen3-coder:480b
- ollama-cloud/deepseek-v4-pro
- ollama-cloud/kimi-k2.6
failover_strategy: mixed
@@ -373,7 +371,6 @@ agents:
- orchestrator
fallback_models:
- ollama-cloud/kimi-k2.6
- ollama-cloud/qwen3-coder:480b
- ollama-cloud/deepseek-v4-pro
failover_strategy: mixed
reasoning_effort: medium
@@ -628,10 +625,10 @@ agents:
produces:
- priority_order
- issue_labels
- issue closures
- issue_closures
forbidden:
- implementation
model: ollama-cloud/kimi-k2.6
model: ollama-cloud/minimax-m2.5:cloud
mode: subagent
delegates_to: []
fallback_models:
@@ -656,7 +653,7 @@ agents:
- code_writing
- code_changes
- prompt_changes
model: ollama-cloud/qwen3-coder:480b
model: ollama-cloud/deepseek-v4-pro
mode: subagent
delegates_to:
- prompt-optimizer
@@ -835,7 +832,7 @@ agents:
forbidden:
- code_changes
- implementation
model: ollama-cloud/qwen3-coder:480b
model: ollama-cloud/deepseek-v4-pro
variant: thinking
mode: subagent
delegates_to:
@@ -846,31 +843,51 @@ agents:
- ollama-cloud/kimi-k2.6
failover_strategy: downgraded
reasoning_effort: low
incident-responder:
history-miner:
capabilities:
- inter_agent_conflict_detection
- architecture_conformance_validation
- state_tracking_sanity
- process_inspection
- uncomfortable_questions_protocol
- pre_flight_validation
- mid_flight_revalidation
- git_history_analysis
- duplicate_detection
- regression_prevention
- pattern_matching
- past_solution_retrieval
receives:
- checkpoint_yaml
- task_claims
- agent_chain
- architecture_docs
- capability_index
- task_description
- codebase_context
produces:
- cross_check_report
- verdict_approved_conditional_blocked
- risk_flags
- mitigation_suggestions
- historical_findings
- regression_warnings
- recommended_solutions
forbidden:
- code_writing
- implementation
model: ollama-cloud/deepseek-v4-pro
mode: subagent
delegates_to: []
fallback_models:
- ollama-cloud/glm-5.1
failover_strategy: downgraded
reasoning_effort: low
incident-responder:
capabilities:
- incident_response
- live_forensics
- malware_removal
- persistence_hunting
- ssh_cleanup
- post_incident_hardening
- cross_platform_hardening
receives:
- incident_report
- server_logs
- threat_indicators
produces:
- forensics_report
- cleanup_actions
- hardening_recommendations
forbidden:
- feature_development
- code_changes
model: ollama-cloud/glm-5.1
mode: subagent
delegates_to:
- orchestrator
- reflector
@@ -881,8 +898,6 @@ agents:
- ollama-cloud/kimi-k2.6
failover_strategy: downgraded
reasoning_effort: high
workflow-cross-checker: null
variant: thinking
evolution-prompt:
capabilities:
- prompt_generation
@@ -906,8 +921,6 @@ agents:
fallback_models:
- ollama-cloud/deepseek-v4-pro
- ollama-cloud/kimi-k2.6
- ollama-cloud/kimi-k2.6
- ollama-cloud/qwen3-coder:480b
failover_strategy: downgraded
reasoning_effort: high
capability_routing:

51
agent-evolution/data/evolution-summary.json
View File

@@ -1,15 +1,15 @@
{
"ts": "2026-06-01T20:35:00Z",
"ts": "2026-06-01T21:30:00Z",
"event": "evolution_complete_report",
"trigger": "user_request_objective_evolution",
"methodology": "capability-analyst_research_report + deterministic_sync",
"agents_changed": 29,
"methodology": "capability-analyst_research_report + deterministic_sync + code_skeptic_review",
"agents_changed": 32,
"model_distribution": {
"deepseek-v4-pro": 14,
"minimax-m3:cloud": 8,
"glm-5.1": 4,
"minimax-m2.5:cloud": 2,
"kimi-k2.6": 1
"deepseek-v4-pro": 16,
"minimax-m3:cloud": 10,
"glm-5.1": 5,
"kimi-k2.6": 5,
"minimax-m2.5:cloud": 3
},
"evidence_file": "agent-evolution/data/research-report.json",
"evidence_sources": [
@@ -21,18 +21,39 @@
"ollama.com/library/minimax-m2.5",
"minimax.io/models/text/m3",
"minimax.io/news/minimax-m25",
"qwenlm.github.io/blog/qwen3-coder"
"qwenlm.github.io/blog/qwen3-coder",
"api.llm-stats.com/v1/ (pricing/provider metadata only, no benchmark scores)"
],
"code_skeptic_findings": {
"issues_fixed": [
"incident-responder in capability-index.yaml had copy-pasted workflow-cross-checker capabilities; replaced with correct incident_response capabilities",
"removed orphaned 'workflow-cross-checker: null' field and unjustified 'variant: thinking' from incident-responder",
"added missing history-miner entry to capability-index.yaml",
"3 model mismatches fixed: product-owner (kimi-k2.6 → minimax-m2.5:cloud), incident-responder (deepseek-v4-pro → glm-5.1), history-miner (qwen3-coder:480b → deepseek-v4-pro)",
"3 additional mismatches fixed: architect-indexer, pipeline-judge, workflow-cross-checker (all qwen3-coder:480b → deepseek-v4-pro)"
],
"total_model_mismatches_fixed": 6
},
"opencompass_container": {
"files": ["docker/docker-compose.opencompass.yml", "docker/Dockerfile.opencompass", "scripts/opencompass-eval.sh", "scripts/opencompass-setup.sh"],
"status": "config_complete_build_blocked_network",
"note": "Docker build requires internet access for pip install. Files validated and ready."
"note": "Docker build requires internet access for pip install. Files validated and ready. Not needed — no benchmark endpoint available."
},
"llm_stats_api": {
"status": "pricing_registry_only",
"benchmarks_available": false,
"models_with_metadata": ["deepseek-v4-pro-max", "glm-5.1", "kimi-k2.6", "minimax-m2.5", "minimax-m2.7"],
"models_not_found": ["minimax-m3", "qwen3-coder-480b"],
"finding": "LLM Stats API (api.llm-stats.com/v1/) provides model registry, pricing, provider metadata, and param_count but has NO benchmark score endpoints. Manual research remains the sole source of benchmark data."
},
"data_gaps": [
"minimax-m3: ALL benchmark tables on ollama.com and minimax.io are IMAGE-ONLY. Specific coding scores unavailable.",
"qwen3-coder-480b: ALL benchmarks image-only. Lowest confidence assignment.",
"kimi-k2.6: Ollama page image-only. Using K2 Instruct as proxy (likely understates performance).",
"minimax-m2.5: Ollama images + partial blog text. Reasoning benchmarks missing."
"minimax-m3: Not found in LLM Stats API. ALL benchmark tables on ollama.com and minimax.io are IMAGE-ONLY. Specific coding scores unavailable.",
"qwen3-coder-480b: Not found in LLM Stats API. ALL benchmarks image-only. No longer assigned to any agent.",
"kimi-k2.6: Ollama page image-only. Using K2 Instruct as proxy (likely understates performance). API provides pricing/providers.",
"minimax-m2.5: Ollama images + partial blog text. Reasoning benchmarks missing. API provides pricing/providers and a 1M context discrepancy (manual said 198K, API shows 1M).",
"minimax-m2.7: Not in manual research. Found in API with release_date 2026-03-18. param_count null in API. SWE-Pro 56.22% from API description."
],
"verification": "scripts/sync-agents.cjs --check PASSED"
"verification": "scripts/sync-agents.cjs --check PASSED",
"total_agents_assigned": 36,
"zero_unassigned": true
}

81
agent-evolution/data/research-report.json
View File

@@ -28,6 +28,83 @@
],
"confidence": "high-for-text-extracted, medium-for-image-only-models"
},
"api_metadata": {
"source": "LLM Stats API (api.llm-stats.com/v1/)",
"fetched_at": "2026-06-01T20:57:00+01:00",
"api_key": "REDACTED",
"endpoints_probed": ["/v1/models", "/v1/models/{id}", "/v1/benchmarks", "/v1/scores", "/v1/rankings", "/v1/evaluations"],
"benchmark_endpoint_status": "NOT_FOUND — API is a pricing/registry API, not a benchmark aggregator",
"models_found_in_api": {
"deepseek-v4-pro-max": {
"param_count": 1600000000000,
"release_date": "2026-04-23",
"description_summary": "DeepSeek-V4-Pro-Max is the maximum reasoning effort mode of DeepSeek-V4-Pro, a 1.6T-parameter MoE model with 49B activated parameters and a 1M-token context window...",
"cheapest_provider": "DeepInfra",
"cheapest_input_price_per_1m": 1.74,
"cheapest_output_price_per_1m": 3.48,
"context_max": 1048576,
"available_in_zeroeval": true
},
"glm-5.1": {
"param_count": 754000000000,
"release_date": "2026-04-07",
"description_summary": "GLM-5.1 is Z.AI's next-generation flagship foundation model designed for long-horizon agentic engineering tasks...",
"cheapest_provider": "FriendliAI",
"cheapest_input_price_per_1m": 1.4,
"cheapest_output_price_per_1m": 4.4,
"context_max": 200000,
"available_in_zeroeval": true
},
"kimi-k2.6": {
"param_count": 1000000000000,
"release_date": "2026-04-20",
"description_summary": "Kimi K2.6 is Moonshot AI's open-source, native multimodal agentic model...",
"cheapest_provider": "Fireworks",
"cheapest_input_price_per_1m": 0.95,
"cheapest_output_price_per_1m": 4.0,
"context_max": 262144,
"available_in_zeroeval": true
},
"minimax-m2.5": {
"param_count": 230000000000,
"release_date": "2026-02-12",
"description_summary": "MiniMax M2.5 is the world's first production-level model designed natively for Agent scenarios...",
"cheapest_provider": "MiniMax",
"cheapest_input_price_per_1m": 0.3,
"cheapest_output_price_per_1m": 1.2,
"context_max": 1000000,
"available_in_zeroeval": true
},
"minimax-m2.7": {
"param_count": null,
"release_date": "2026-03-18",
"description_summary": "MiniMax M2.7 features model self-improvement driving productivity innovation...",
"cheapest_provider": "Fireworks",
"cheapest_input_price_per_1m": 0.3,
"cheapest_output_price_per_1m": 1.2,
"context_max": 204800,
"available_in_zeroeval": true
}
},
"models_not_found": [
{ "model": "minimax-m3", "reason": "No entry in API catalog. All data from manual research." },
{ "model": "qwen3-coder-480b", "reason": "No entry in API catalog. Only smaller Qwen3 variants present. All data from manual research." }
],
"discrepancies_with_manual_research": [
{
"field": "minimax-m2.5.context",
"manual_value": "198K tokens",
"api_value": "1000000 (MiniMax provider)",
"verdict": "API shows 1M context for MiniMax provider, manual said 198K. Likely manual refers to a different provider or older spec."
},
{
"field": "minimax-m2.7.param_count",
"manual_value": "not researched (not in manual report)",
"api_value": "null",
"verdict": "API does not provide param_count for M2.7."
}
]
},
"models": {
"deepseek-v4-pro": {
"vendor": "DeepSeek",
@@ -416,13 +493,13 @@
},
"planner": {
"best_model": "minimax-m3",
"rationale": "PostTrainBench #3 demonstrates autonomous planning + execution. 12h autonomous tasks. 300-agent swarm coordination. Best for complex task decomposition.",
"rationale": "PostTrainBench #3 (37.1) demonstrates autonomous planning + execution. 12h autonomous tasks (ICLR replication, 18 commits). Best for SOLO deep task decomposition. CRITICAL CORRECTION: '300-agent swarm' claim belongs to kimi-k2.6 (unverified marketing prose); minimax-m3 has ZERO verified multi-agent swarm capability.",
"fallback": "glm-5.1 (sustained multi-round planning without plateauing)"
},
"orchestrator": {
"best_model": "glm-5.1",
"rationale": "UNIQUE CLAIM: sustained performance over hundreds of rounds, thousands of tool calls. Does not plateau. Designed for agentic engineering. Vending Bench $5,634 (economic task competence).",
"fallback": "minimax-m3 (agent swarm coordination, 12h autonomous runs)"
"fallback": "minimax-m3 (solo long-horizon autonomous tasks, 12h autonomous runs)"
},
"agent-architect": {
"best_model": "minimax-m3",

14
kilo-meta.json
View File

@@ -1,7 +1,7 @@
{
"$schema": "https://app.kilo.ai/config.json",
"metaVersion": "1.0.0",
"lastSync": "2026-06-01T19:50:01.425Z",
"lastSync": "2026-06-01T21:00:25.859Z",
"agents": {
"requirement-refiner": {
"file": ".kilo/agents/requirement-refiner.md",
@@ -14,7 +14,7 @@
"history-miner": {
"file": ".kilo/agents/history-miner.md",
"description": "Analyzes git history to find duplicates and past solutions, preventing regression and duplicate work",
"model": "ollama-cloud/qwen3-coder:480b",
"model": "ollama-cloud/deepseek-v4-pro",
"mode": "subagent",
"category": "core"
},
@@ -145,7 +145,7 @@
"product-owner": {
"file": ".kilo/agents/product-owner.md",
"description": "Manages issue checklists, status labels, tracks progress and coordinates with human users",
"model": "ollama-cloud/kimi-k2.6",
"model": "ollama-cloud/minimax-m2.5:cloud",
"mode": "subagent",
"category": "meta"
},
@@ -211,7 +211,7 @@
"architect-indexer": {
"file": ".kilo/agents/architect-indexer.md",
"description": "Indexes and maps project codebase architecture into .architect/ directory",
"model": "ollama-cloud/qwen3-coder:480b",
"model": "ollama-cloud/deepseek-v4-pro",
"mode": "subagent",
"color": "#10B981",
"category": "core"
@@ -235,7 +235,7 @@
"pipeline-judge": {
"file": ".kilo/agents/pipeline-judge.md",
"description": "Automated pipeline judge. Evaluates workflow execution by running tests, measuring token cost and wall-clock time. Produces objective fitness scores. Never writes code - only measures and scores.",
"model": "ollama-cloud/qwen3-coder:480b",
"model": "ollama-cloud/deepseek-v4-pro",
"mode": "subagent",
"color": "#DC2626",
"category": "meta"
@@ -251,7 +251,7 @@
"incident-responder": {
"file": ".kilo/agents/incident-responder.md",
"description": "Server incident response and system hardening specialist. Handles live forensics, malware removal, persistence hunting, SSH-based server cleanup, and post-incident hardening. Works with any OS and panel.",
"model": "ollama-cloud/deepseek-v4-pro",
"model": "ollama-cloud/glm-5.1",
"mode": "subagent",
"color": "#B91C1C",
"category": "core"
@@ -259,7 +259,7 @@
"workflow-cross-checker": {
"file": ".kilo/agents/workflow-cross-checker.md",
"description": "Workflow cross-checker and process inspector. Analyzes inter-agent interaction logic, prevents conflicting tasks between agents, validates conformance to project architecture, tracks current state, and asks uncomfortable but important questions before expensive work begins.",
"model": "ollama-cloud/qwen3-coder:480b",
"model": "ollama-cloud/deepseek-v4-pro",
"mode": "subagent",
"color": "#9333EA",
"category": "meta"

12
kilo.jsonc
View File

@@ -43,7 +43,7 @@
"history-miner": {
"description": "Analyzes git history to find duplicates and past solutions, preventing regression and duplicate work",
"mode": "subagent",
"model": "ollama-cloud/qwen3-coder:480b",
"model": "ollama-cloud/deepseek-v4-pro",
"permission": {
"task": {
"*": "deny",
@@ -358,7 +358,7 @@
"product-owner": {
"description": "Manages issue checklists, status labels, tracks progress and coordinates with human users",
"mode": "subagent",
"model": "ollama-cloud/kimi-k2.6",
"model": "ollama-cloud/minimax-m2.5:cloud",
"permission": {
"read": "allow",
"edit": "allow",
@@ -502,7 +502,7 @@
"architect-indexer": {
"description": "Indexes and maps project codebase architecture into .architect/ directory",
"mode": "subagent",
"model": "ollama-cloud/qwen3-coder:480b",
"model": "ollama-cloud/deepseek-v4-pro",
"color": "#10B981"
},
"flutter-developer": {
@@ -520,7 +520,7 @@
"pipeline-judge": {
"description": "Automated pipeline judge. Evaluates workflow execution by running tests, measuring token cost and wall-clock time. Produces objective fitness scores. Never writes code - only measures and scores.",
"mode": "subagent",
"model": "ollama-cloud/qwen3-coder:480b",
"model": "ollama-cloud/deepseek-v4-pro",
"color": "#DC2626"
},
"python-developer": {
@@ -532,7 +532,7 @@
"incident-responder": {
"description": "Server incident response and system hardening specialist. Handles live forensics, malware removal, persistence hunting, SSH-based server cleanup, and post-incident hardening. Works with any OS and panel.",
"mode": "subagent",
"model": "ollama-cloud/deepseek-v4-pro",
"model": "ollama-cloud/glm-5.1",
"color": "#B91C1C",
"permission": {
"read": "allow",
@@ -552,7 +552,7 @@
"workflow-cross-checker": {
"description": "Workflow cross-checker and process inspector. Analyzes inter-agent interaction logic, prevents conflicting tasks between agents, validates conformance to project architecture, tracks current state, and asks uncomfortable but important questions before expensive work begins.",
"mode": "subagent",
"model": "ollama-cloud/qwen3-coder:480b",
"model": "ollama-cloud/deepseek-v4-pro",
"color": "#9333EA",
"variant": "thinking",
"permission": {