AI & ML

sohnajonin Skill

Generate production-ready static ad images for any brand using Claude + Nano Banana 2. End-to-end automated workflow from brand research → prompt generation → image generation via FAL API. Trigger on requests to create static ads, generate ad creatives, build ad images, or when user mentions Nano Banana, Higgsfield, FAL, or static ad generation. Also trigger when user drops a brand name + URL and asks for ad creatives.

reinforce52 Bundle

个人镜像（Personal Twin）：构建并持续维护一份关于"用户本人"的结构化画像，让 OpenClaw 能针对用户的知识水平、学习风格、性格倾向和易错点提供个性化服务。三层画像：①静态档案 （基本情况/目标）②证据驱动的能力熟悉度画像（不做严格 BKT 知识追踪，只依据笔记/项目/对话/ 错题卡点做 0-4 级熟悉度估计）③认知与人格画像（大五 BFI 量表打底 + 学习风格 VARK + AI 长期 微调，人格以量表为锚、不做心理诊断）。画像目录固定为 E:\obsidian\rein\_个人镜像\，每次更新 生成精简《_注入摘要.md》并同步到 openwave self-model，实现每轮自动"懂你"。 触发指令：【个人镜像 初始化】（首次：建档案+做量表）、【记录卡点】<一句话问题/错误/解决>、 【个人镜像 刷新】（立即全量重算）、【我现在什么水平】（读取画像出分析）、【更新档案】。 更新策略：默认静默自动更新（Windows 任务计划 PersonalTwin-Daily 每日23:00增量、 PersonalTwin-Weekly 周日21:00全量；用系统计划而非 smart-cron，电脑开机即跑、不依赖 OpenClaw 网关在线）， 每次留快照与 changelog， 用户随时可查 04_成长轨迹.md。红线：只扫描指定 Obsidian 库；熟悉度必须带证据与时间、不编造掌握； 人格以量表为锚、AI 只能补充不能被一两次对话推翻、不贴病理标签；自动更新留快照可回滚。

reinforce52 Bundle

一站式求职自动化工具集：多渠道职位搜索（BOSS直聘/中南大学就业网/实习僧）+ AI匹配打分 + 基于Obsidian知识库的定制简历生成（Drafter-Reviewer双Agent循环 + JD预分析8维度报告）+ 技能差距学习路径（解析JD缺失技能→生成学习计划→调用topic-knowledge-base建库→学完更新简历）+ PDF/Word/Markdown多格式导出。支持一键流程：搜索→匹配→选岗→生成简历→技能差距分析。 典型触发语：'生成简历 方向=xxx'、'找工作 关键词=xxx'、'一键求职 xxx'、'更新简历'、 '帮我做一份xxx岗位的简历'、'搜索xxx岗位'、'匹配度分析'、'帮我分析缺什么技能'、 '技能差距学习'、'学完更新简历'。输出到 E:\obsidian\rein\简历库\。

reinforce52 Bundle

将 OpenClaw 对话路由到最佳中文大模型（DeepSeek/Qwen/GLM/Kimi/Doubao/MiniMax/Step/Baichuan/Spark/Hunyuan）， 无需复杂配置即可切换模型。触发词：'list models'、'use <model>'、'compare <models>'、'哪个中文模型最适合编程/长文本/视觉，切换到它'。 当用户询问中文模型选择、切换或比较能力与价格时使用。

reinforce52 Bundle

把 YouTube / B站 / 抖音 视频自动转化为 Obsidian 深度学习知识库。 v2 升级：细粒度知识点拆分 + web深度搜索扩展 + 三信源融合（视频+web+可选教材）+ 9段式深度笔记 + 知识谱系总图（Mermaid mindmap）+ 学习闭环（复习清单+掌握检验）。 工作流：用户贴视频链接 → 自动下载(字幕优先/无字幕转写) → LLM细粒度知识点拆分（5-15个）→ 每知识点4-6路web深度搜索 → 三信源融合 → 视频总览笔记 + 每个知识点独立9段式笔记 → 知识谱系总图 → 复习清单。支持单视频和批量(播放列表/频道)。 典型触发语：'把这个视频整理成笔记'、'视频转知识库'、'总结这个YouTube/B站/抖音视频'、 '学习这个视频'、'视频笔记 <链接>'、'批量整理这个播放列表'、'把这个频道的视频都做成笔记'。

sharp-skills Bundle

Bound agent-to-agent delegation with explicit budgets — max recursion depth, max fan-out per agent, max total agents, and cycle detection — so a system where agents spawn subagents can't explode into a runaway swarm. Use this when agents can delegate to or spawn other agents (orchestrator-subagent, swarm, recursive planning), when a run's cost or latency grows unpredictably, or when you suspect a delegation loop. Depth and fan-out each look reasonable locally while their product is a bomb.

sharp-skills Bundle

Decide how much authority one agent's message carries to another by the channel it arrived on, never by the source it claims — so a spoofed or injected instruction can't escalate scope or disable safety just by claiming to come from "the orchestrator." Use this when agents send each other instructions, when building orchestrator-subagent or swarm architectures, or when an agent acts on tool output, forwarded messages, or "another agent said." Instructions from another agent deserve no more authority than instructions from an unknown human.

sharp-skills Bundle

Get a trustworthy typed value out of a model's response, or a clean re-ask — never a guess. Two shapes: free-form prose you can't get as JSON (extract a verdict/score/boolean, and flag conflicting values as ambiguous instead of grabbing the first match), and "mostly JSON" wrapped in fences (extract, safe-repair, validate against a schema, re-ask on failure). The rule both share: never invent a value. Use when a downstream agent consumes another agent's output and a naive regex or `json.loads` returns the wrong value or crashes. Works on any model's text, not a provider format.

sharp-skills Bundle

Configure the context compaction that modern agent runtimes already provide, instead of coding your own — and get the settings right so it shrinks history without destroying it. Use this when long or multi-agent runs overflow the model's context window, when you're choosing which model should summarize old turns, or when compaction is "on" but the agent keeps losing track of things. The load-bearing rule: the summarizer's context window must be at least as large as the main model's, or the compaction that was supposed to save the run silently corrupts it.

sharp-skills Bundle

Validate an agent's proposed tool calls before they execute — known tool, complete arguments, destructive operations explicitly approved, no redundant repeats — so a model's mistake can't become a real side effect, and shrink what needs validating by pinning the parameters that never legitimately vary instead of delegating them all to the model. Use this when an agent can run shell commands, write files, call APIs, or otherwise act with consequences; when you see loops of the same call fired repeatedly; when a tool ran on a placeholder path the model never filled in; or when defining a tool's parameters. Works for any tool-using runtime (Claude Code, OpenAI/Codex function calls, LangGraph/CrewAI tool nodes) — it inspects the call, not the model.

sharp-skills Bundle

Keep an agent's persistent memory trustworthy — every entry has provenance, entries expire and get re-verified, secrets never land in memory, and untrusted content is never stored as fact — so recall stays reliable instead of slowly poisoning the agent. Use this when agents write to and read from a long-lived memory store, when old remembered "facts" start misleading current runs, when memory grows without bound, or when you worry a prompt injection could persist into memory. The recall a stale or poisoned memory returns is worse than no memory at all.

bizpers11991-code Skill

Render animations from Blender headlessly via a Python scene description - EEVEE setup, the Blender 4.4/5.x API breaks (slotted actions, compositor node groups, glare sockets), camera orientation maths, and the object-count cost model that decides whether a scene builds in 30 seconds or 6 minutes.

sharp-skills Bundle

Keep prompt injection out of a model by making the untrusted-vs-instruction boundary structural and unescapable, not by regex. Enforce that untrusted content (web, inbox, tool output) is delimited as data, and — the load-bearing check — that it can't escape the boundary by containing its own delimiter; markers must be an unpredictable per-item nonce. Injection signatures are flagged too, but as a bypassable second wall behind delimiting, isolation, and message-trust. Use when an agent ingests content an attacker could author. Inspects content, not a provider format.

sharp-skills Bundle

Gate an agent on a red-team suite that tries to make it misbehave — jailbreaks, injected instructions, scope escalations, harmful requests — each with the safe behavior it must show, scored objectively before it ships. Use this when an agent is exposed to adversarial input or acts with real consequences, when you need a repeatable security regression test rather than ad-hoc probing, or when "it seems safe" isn't evidence. Distinct from quality review and completeness checks: this asks whether the agent can be *made to fail*, not whether its output is good.

praisondani Bundle

Search public RecipeBee recipes and save private recipes or meal plans through RecipeBee's public API and MCP server. Use when the user wants to find a recipe, preview or save a recipe in RecipeBee, or import a meal plan.

reinforce52 Bundle

分析各类竞赛/比赛并指导制备参赛交付物。当用户提出"分析比赛/了解比赛/比赛需要什么技能/比赛知识储备/比赛选题/写项目申报书/比赛Demo或演示视频/技术架构图/现场演示/比赛源码仓库/找比赛可用的GitHub开源项目/工程方向agent开发"等需求时使用。覆盖创新创业类（挑战杯、互联网+）与AI智能体/大模型应用类赛事（如"海之子"杯AI智能体挑战计划），输出：①比赛规则、赛程与评分解读 ②所需技能与知识储备清单 ③选题方向分析与落地思路 ④五大交付物（项目申报书、≤5分钟Demo/演示视频、1页技术架构图、现场演示、完整源码仓库）制备指引 ⑤GitHub成品级开源项目检索与复用建议。

reinforce52 Bundle

当用户发送【同步Obsidian 项目路径=xxx】或【追加补充学习】时使用，作为学习汇总中枢 Agent，把每日学习 （VSCode-Harness 的 daily_summary.md + 最近24h git 提交 + 会话补充）整合成 Obsidian 分层知识库 （每日日志/学科学习库/项目库/简历素材库，Vault = E:\obsidian\rein）。 核心约束：写前完整预览并等用户【确认】；绝不静默写入、绝不覆盖已有笔记、绝不虚构学习成果、绝不读 VSCode-Harness 内部对话。能力：当天日志重复检测、容错占位、backfill 补任意日期、每日复习题、git log --stat。 典型触发语：'同步Obsidian 项目路径=D:/cv-projects/civil-yolo-detect'（可加 日期=YYYY-MM-DD）、'追加补充学习'、 '把今天的学习同步到Obsidian'。工作流：解析路径 → 读 daily_summary.md → 读24h git log（只读）→ 收集会话补充 → 按四大库模板生成全部文档 → 强制输出完整预览+文件路径清单 → 等【确认】→ 文件直写 E:\obsidian\rein → 写后回读校验 → 输出完成报告。

reinforce52 Bundle

当用户提供一个资料包/文件夹路径，希望把其中的 PDF、Word、Markdown 等资料阅读分析后，按"先总后细"的知识谱系 整理成 Obsidian 知识库（含知识谱系图、复习闭环、知识关联、增量日志与写后回读校验）时使用。 典型触发语：'帮我整理这个文件夹里的资料'、'把这些 PDF/Word 整理成知识库'、'分析这个资料包并建学习笔记'、 '把 D:\\xxx 整理成 Obsidian 笔记'。 工作流：确认资料路径 → 查已有库/资料指纹去重 → 扫描预检 → 逐份提取文本（含 OCR 管线）→ LLM 识别主题、 多级分类与知识谱系树（先总后细）→ 按谱系树建目录并生成 _知识谱系.md 与各 _总览.md → 生成结构化笔记 （来源/前置知识/关联知识点/自我检测/复习排期）→ 生成 _MOC 与 _复习清单 → 写后回读校验 → 汇报。 每个主题为独立 vault（E:\obsidian\rein\<主题>\），不统一路径；只读资料包，不修改原文件。

sharp-skills Bundle

Drive a multi-agent event-driven system from one machine-readable registry and have the runtime derive routing from it instead of duplicating it. Use when agents or events are defined in more than one place (config, hardcoded graph, docs) and they have drifted; you see orphan subscriptions (an agent subscribes to an event no agent publishes); you hit dead-end publishes (an event emitted into the void); or your bus routing graph is out of sync with the registry.

sharp-skills Bundle

Define one wire contract for every event that crosses the bus — required correlation fields, a runtime tag, a mode tag — so a single control plane can re-enter a run identically no matter which runtime (local model, remote worker, browser, or plain code) emitted the event. Use this when agents run on more than one runtime and you need cross-runtime tracing and routing to stay consistent, when adding a new runtime keeps breaking event handling somewhere downstream, or when you want envelope validation that runs the same everywhere without a schema-library dependency.

sharp-skills Bundle

Keep a prompt-injected or mistaken agent from doing real damage by isolating what each session can touch — never letting one session hold sensitive access, untrusted input, and an outbound channel at the same time (the lethal trifecta). Use this when an agent logs into real accounts, drives a browser, or reads inbound messages/web pages an attacker can author; when deciding which capabilities may share a session; or when giving an agent bounded autonomy over money, publishing, or communication. A sandbox or VM protects your host and network — it does not protect the accounts you log in inside it.

sharp-skills Bundle

Let an agent earn the right to act instead of being configured with it — a four-rung ladder from observe to act, promotion paid for by a trailing run of clean human verdicts that one edit resets, grants keyed on (action, resource, principal), and the invariant that an agent can never grant itself. Choose the rung by whether a human actually watches the review channel, not by how important the action feels. Use when deciding what an agent may do without asking, when approval fatigue is pushing a team toward blanket permissions, or when permissions were set once at integration time and never revisited.

sharp-skills Bundle

Give a multi-agent system persistent experience — per-agent smoothed confidence and structured learnings recorded at run completion, loaded at session start, and periodically distilled into prompt/skill improvements. Use this when every session starts from zero and repeats last month's mistakes, when agent performance data exists only in people's memory, when accumulated "learnings" JSON grows but nothing changes, or when you need to know which agents are trending down.

om-talaviya Bundle

Converts API curl commands from documentation into fully functional Claude Agent Skills (complete with natural-language triggers, YAML frontmatter, parameter documentation, and a standalone executable runner script). Use whenever the user provides a curl command, asks to wrap a curl request into a skill, or asks to generate an agent skill from an API endpoint.

om-talaviya Skill

Converts API curl commands from documentation into fully functional Claude Agent Skills (complete with natural-language triggers, YAML frontmatter, parameter documentation, and a standalone executable runner script). Use whenever the user provides a curl command, asks to wrap a curl request into a skill, or asks to generate an agent skill from an API endpoint.

om-talaviya Bundle

Guides Indian exporters (goods or services) through GST compliance, high-risk regulatory traps, and refund mechanics — LUT (Letter of Undertaking) filing, zero-rated supply treatment, Rule 96(10) Advance Authorisation restrictions, Rule 96B foreign exchange non-realization clawback, IGST refund via ICEGATE/shipping bill, GSTR-1 and GSTR-3B reconciliation, Export of Services vs. Intermediary tax traps (Section 13(8)(b)), and RFD-01 refund applications. Use this whenever the user mentions exporting from India, GST on exports, LUT, zero-rated supply, IGST refund, shipping bill, ICEGATE, GSTR-1, GSTR-3B, RFD-01, EDPMS, FEMA 9-month realization, or asks how GST works for their export/import business. This is a preparer/explainer skill, not a filing agent — it produces computations, checklists, and draft data, and always tells the user to verify current rates/forms and consult a CA before filing.

yihuil1992 Bundle

Land GitHub pull requests after they are ready. Use when an AI coding agent is asked to wait for CI/checks, confirm review state, merge a PR, delete the PR branch, switch back to main/master/base branch, and pull the latest changes after merge. Also use for requests like 'merge this PR after CI', 'land the current PR', 'ship this branch', 'merge and clean up', or Chinese requests that ask to merge after CI and clean up branches.

yihuil1992 Bundle

Use when an AI coding agent should turn a complex or fuzzy request into a bounded implementation workflow: clarify requirements without overwhelming the user, bootstrap local AGENT/SYSTEM_MAP/WIP/ADR docs when missing, draft and self-review a task spec from fresh perspectives, converge it to approval, implement it with the current agent by default, verify the work, and preserve follow-up memory in SYSTEM_MAP/ADR/AGENT or a checklist. Trigger on requests like '走 spec 流程', '先 brainstorm', '先写 spec', '自己审到 approved', '需求比较复杂', '等着验收', or when repository AGENT(S), docs/wip, agent-docs/wip, ADR, SYSTEM_MAP, or local project-memory workflows are involved.

riclyme Skill

Builds publication-quality Word (.docx) regression tables for strategy/management journals (SMJ, JMS, AMJ, ASQ style) using python-docx. Covers SMJ manuscript conventions (double-space body, APA headings, tables at end), standard model progression (M1=controls, M2=IV+controls, M3+=moderator+IV+interaction+controls), compact β/[p]/(SE) cell format, academic top-bottom borders, landscape section breaks for wide tables, merged moderator rows, FE as "Yes", VIF reporting, and correlation+descriptive statistics tables (numbered lower-triangle with Mean/SD rows). Use when building or reformatting any regression, correlation, or descriptive statistics table in Word.

sharp-skills Bundle

Make a long-running agent session survive being interrupted — log every action so recovery replays recorded outcomes instead of re-running them, key side effects so they cannot fire twice, bound every retry, and scope caches so one user's answer never becomes another's. Use when a session lives longer than the process running it (hours, days, weeks), when a crash or restart forces users to start over or answer the same question again, when a retry loop burns budget without ever escalating, or when you need the production-failure axis rather than the output-quality one.

sharp-skills Bundle

Enforce multi-agent governance rules — no-touch zones, secret access, config drift — in the harness via hooks, instead of asking the model to police itself in prompts. Use this when an agent edited a file it was told never to touch, when prompt rules like "never read .env" keep being violated under context pressure, when you need an auditable block rather than a polite instruction, or when governed config files change without review.

sharp-skills Bundle

Keep a multi-agent system's agent prompts, output JSON schemas, and routing registry aligned as one enforced contract — no layer silently teaches or accepts a status the router will reject — and keep each prompt under real change control: pinned by content hash, changelogged, rollback-ready. Use this when a run dies in a reject/retry loop on a "valid-looking" output, when schemas or prompts drift from the registry, when a prompt's reference example contradicts the schema, before flipping a contract from warn to enforce, or when a prompt edited in place makes a regression untraceable to a version.

securitytalent Skill

Intelligent orchestration layer that analyzes requests, selects the most relevant OpenCode skills, combines workflows intelligently, and coordinates multi-skill execution.

securitytalent Bundle

Comprehensive cybersecurity tabletop exercise design and facilitation framework. USE WHEN designing incident response scenarios, creating executive or technical tabletops, generating atomics for exercise runners, identifying missing SOPs/playbooks, or evaluating organizational preparedness. Includes threat model integration, CISA-aligned methodologies, and automated gap analysis.

securitytalent Bundle

Pulse - Daily Curated News Digest

securitytalent Bundle

Web cache deception and poisoning exploitation. USE WHEN user mentions cache deception, cache poisoning, CDN bypass, URL parsing discrepancy, path confusion, static extension bypass, or cache key manipulation. Based on Martin Doyhenard's "Gotta Cache 'em all" research.