deepseek-v3-lite-agent

Transform AI agents from task-followers into proactive partners with memory architecture, reverse prompting, and self-healing patterns. Lightweight version focused on core proactive capabilities.

Ultimate AI agent memory system with WAL protocol, vector search, git-notes, and cloud backup. And also 50+ models for image generation, video generation, text-to-speech, speech-to-text, music, chat, web search, document parsing, email, and SMS.

Integrates DeepSeek API with OpenClaw agents.

Build, harden, and scale elite Discord servers with a practical admin playbook: security baseline, role/permission architecture, onboarding, moderation ops, engagement systems, and analytics-driven iteration. Use when designing a new server, auditing an existing one, fixing chaos, or preparing for growth.

Agente especialista em Literatura Brasileira e Portuguesa para FUVEST — obras obrigatórias 2026 e movimentos literários. Opera exclusivamente no Modo Conversa: as obras não são estudadas, são vividas — através de personagens reais, contextos históricos palpáveis e conexões com filosofia, sociologia e a vida de Antonio. As 9 obras obrigatórias da FUVEST 2026 são todas de autoras mulheres: Clarice Lispector, Lygia Fagundes Telles, Rachel de Queiroz, Conceição Evaristo, Sophia de Mello Breyner Andresen, Paulina Chiziane, Djaimilia Pereira de Almeida, Julia Lopes de Almeida, Narcisa Amália. Acionar quando Antonio mencionar qualquer obra, autora, movimento literário, quiser conversar sobre leitura, ou o orchestrator detectar oportunidade de conexão com Literatura. Funciona nos intervalos (prazer) e integrado ao bloco de Português (análise).

使用 DeepSeek 网页版进行互联网查询，分担大模型请求和搜索负担。当用户需要查询最新信息、一般性知识、代码问题、文本分析等，或明确说"用 DeepSeek 查一下"、"联网搜索"、"查下最新"等时触发此技能。特别地，如果提问以"ds:"或"ds："开头，优先使用此技能。通过 Chrome DevTools MCP 控制浏览器与 DeepSeek 交互，自动检测登录状态并提示用户，保持浏览器会话复用，使用 evaluate_script 提取 Markdown 内容直接返回。

Parse, extract text from, and screenshot PDF and document files locally using the LiteParse CLI (`lit`). Use when asked to extract text from a PDF, parse a Word/Excel/PowerPoint file, batch-process a folder of documents, or generate page screenshots for LLM vision workflows. Runs entirely offline — no cloud, no API key. Supports PDF, DOCX, XLSX, PPTX, images (jpg/png/webp), and more. Triggers on phrases like "extract text from this PDF", "parse this document", "get the text out of", "screenshot this PDF page", or any request to read/extract content from a file.

A medical-research-native literature reading skill for users with clinical, bioinformatics, translational, and basic experimental backgrounds. Use this skill whenever a user wants to read, analyze, critique, or interpret a medical or scientific paper — whether they provide a PDF, abstract, DOI, PMID, or just a title. Triggers include requests like "analyze this paper", "critique this study", "is this a strong paper?", "give me similar studies", "prepare me for journal club", "help me understand this bioinformatics paper", "what are the weaknesses here?", or "turn this into a mind map". Also activate for any downstream deliverables such as journal club kits, comparison tables, PI decision briefs, replication starters, or follow-up experiment designs. Do NOT treat as a generic summarizer — this skill performs structured evidence-type classification, track-specific critical appraisal, interpretation-boundary judgment, and research-grade follow-up generation.

Direct access:

对 AI 自动化流程做轻量红队演练，聚焦误用路径、边界失败和数据泄露风险。；use for red-team, ai, workflow workflows；do not use for 输出可直接滥用的攻击脚本, 帮助破坏系统.

One-click model liberation toolkit for removing refusal behaviors from LLMs via surgical abliteration techniques

Expert skill for using DeepSeek-OCR, a vision-language model for optical character recognition with context optical compression supporting documents, PDFs, and images.