Skills
#hallucination
SkillCategoryDownloads
MARL — Multi-stage Reasoning Middleware@cutechicken99
#multi-agent#middleware
Multi-stage multi-agent reasoning middleware that reduces LLM hallucination…
Lifestyle, Agents
Updated 6mo ago 6 1.4k
Anti-Hallucination@tooled-app
#agent#hallucination
Detects and mitigates hallucinations in agent outputs by self-checking facts,…
Agents
Updated 2mo ago 1 1.3k
Hallucination Guard@reikys
#guard#hallucination
Execution-based verification guardrail with 14 check items for AI agent output
Other
Updated 6mo ago 0 992
Rag Hallucination Governor@william202404
#rag#hallucination
RAG幻觉治理工具。输入RAG系统当前指标/症状/检索日志片段,输出根因诊断+阈值调参建议+治理方案。基于10+生产RAG交付实战经验,不是通用RAG教程。
Knowledge, Operations
Updated 4mo ago 0 1.1k
幻觉检测器@534422530
#hallucination#代码审查
幻觉检测器 - 原创技能。检测AI生成代码中的幻觉内容,包括不存在的API、错误的参数、虚假引用等。适用于代码审查、AI编程辅助、质量保证等场景。
Development
Updated 6mo ago 0 827
Verificate Cn@verificate-dev
#code-review#validation
OpenClaw 的信任层。在信任之前,先验证 AI 编写的代码、工具调用与研究答复——17 道确定性现实闸门 +…
Development, Integrations, Security
Updated 3mo ago 0 459
Verificate@verificate-dev
#code-review#validation
The trust layer for OpenClaw. Verify AI-written code, tool calls and research…
Security, Development, Integrations
Updated 3mo ago 0 458
verify-conclusion@rocketdemon
#verification#reasoning
诊断结论、调研发现或推断出炉后就地开始自检:先逐环拆推理链,再换思路从零盲推;审查对象是计划/方案/变更单时第一轮就先建外部完备性基线再逐维对照,拦截过早结论…
Agents, Development
Updated 5d ago 0 242
