| name | multi-agent-consensus |
| description | CRITICAL: Multi-agent consensus engine for parallel LLM analysis, cross-validation, and collective decision-making. Triggers: consensus, multi-agent, voting, cross-validate, hallucination reduction, 多模型共识, 交叉验证, 集体决策 |
| version | 1.0.0 |
| updated | 2025-02-07T00:00:00.000Z |
| author | ssszyy |
| tags | ["consensus","multi-agent","parallel-execution","voting","stategraph","hallucination-reduction","smart-contract-audit","cross-validation"] |
| triggers | [{"type":"keyword","keywords":["consensus","multi-agent","voting","parallel analysis","cross-validate","collective intelligence","hallucination","multi-model","smart contract audit","security audit"],"priority":90},{"type":"pattern","patterns":["(?i)(analyze|audit|review) .*(with|using) .*multiple .*(model|agent|llm)","(?i)(cross|multi).*(validate|verify|check)","(?i)(reduce|prevent|detect) .*hallucination","(?i)(consensus|vote|agree) .*(decision|result|analysis)","(?i)(parallel|concurrent) .*(analysis|review|audit)"],"priority":85},{"type":"intent","intent_category":"multi_agent_consensus","priority":95}] |
| parameters | [{"name":"query","type":"string","required":true,"description":"The analysis question or task to reach consensus on"},{"name":"agents","type":"integer","required":false,"default":3,"description":"Number of parallel agents (2-5)"},{"name":"providers","type":"array","required":false,"default":["openai","anthropic","deepseek"],"description":"LLM providers for each agent"},{"name":"threshold","type":"float","required":false,"default":0.6,"description":"Minimum agreement ratio to reach consensus (0.0-1.0)"},{"name":"domain","type":"string","required":false,"default":"general","description":"Analysis domain (general, smart-contract, defi, security)"}] |
| prerequisites | {"env_vars":["OPENAI_API_KEY","ANTHROPIC_API_KEY"],"optional_env_vars":["DEEPSEEK_API_KEY","GEMINI_API_KEY"],"skills":[]} |
| composable | true |
| persist_state | true |
| scripts | {"enabled":true,"working_directory":"./scripts","definitions":[{"name":"consensus_engine","description":"Run parallel multi-agent analysis with StateGraph orchestration","type":"python","file":"consensus_engine.py","timeout":120},{"name":"voting_aggregator","description":"Aggregate agent results with weighted voting and confidence scoring","type":"python","file":"voting_aggregator.py","timeout":30}]} |
