Compare pasted decision records and meeting notes to identify conflicting choices, missing context, and unresolved questions while retaining a traceable source ledger.
Turn video scripts into structured retention audits — map hook failures, desertion windows, and re-engagement fixes using attention psychology.
LLMエージェントがセッションをまたいで記憶を保持するための永続メモリ層。抽象基底クラス MemoryStore ＋ LocalJSONStore 実装で、保存・取得・検索を再利用可能に。Karpathy の「LLM OS（記憶＝ディスク層）」着想。バックエンド差し替え可・Python標準ライブラリのみ・LLM呼び出しゼロ・ユニットテスト同梱。
You write one great idea — then lose hours reformatting it for blog, email, X, YouTube, and Instagram. Content Repurposer fixes that: paste one memo, outline, or transcript and get five platform-ready outputs in one run — blog with 10 headlines, 10 X posts, newsletter with subject lines, YouTube script, and 10-slide Instagram carousel. Works with Claude Code, Codex, Cursor, and other AI agents. No API keys or integrations. Anti-hallucination rules keep every output faithful to your source.
I've spent years building production software, designing enterprise architectures, reviewing thousands of code changes, and optimizing AI-assisted development workflows. This Skill packages proven engineering, planning, documentation, testing, and code-review practices into a lightweight workflow that helps AI understand any project before making changes, reducing mistakes, hallucinations, unnecessary token usage, and repeated context loading.
批量抓取下载视频 转换MD文件形式 存入知识库并建知识关联 自我检查 审核 修复 约束 推进 适合长线任务 准备条件 安装LLM wiki Obsidian 知识库 有一个多模态的模型或者本地模型 此skill 来源于Hermes agent反复测试 其他agent 也适用 也可以改造成你自己个性化流程 需要的脚本一并打包上传 需要其他配置智能体自动安装
Your agent or chatbot keeps getting document content wrong, citing passages that don't exist, or saying "I can't find it" about things plainly in the docs. This reframes it as a retrieval failure, not a prompt bug: corpus profile, chunking, embedding and vector-DB index choice, retrieval strategy (top-k, hybrid BM25 + dense, rerank, query rewrite), context assembly with citations, and a golden Q&A set scored on recall@k / MRR / faithfulness.
Research Context Management helps agents run long, evidence-heavy research without losing sources, claims, contradictions, or next steps. It builds a durable workspace with a source ledger, batch notes, claim matrix, contradiction register, working brief, and handoff checkpoint, so deep research, audits, market scans, and reports can be reviewed, resumed, and turned into reliable outputs with clearer evidence trails.
Created from 10+ years of manufacturing and industrial automation experience across food packaging, planter-part manufacturing, and Tier 1 automotive environments, with Ignition used in production-facing work over that span. The skill packages practical SCADA judgment around maintainable expressions, operator-safe behavior, troubleshooting habits, and realistic Designer/Gateway validation workflows.