AI & ML

bkjohn2018 Skill

Assesses whether an AI agent or agentic workflow is ready for finance and accounting use based on autonomy, tool access, data sensitivity, control impact, human approval, failure modes, logging, monitoring, rollback, evidence, and recertification. Use before deploying, approving, expanding, or materially changing an AI agent.

bkjohn2018 Skill

Metric Governance

bkjohn2018 Skill

Data Quality Controls

bkjohn2018 Skill

Designs data security and privacy controls for sensitive, regulated, financial, and business-critical data using NIST SP 800-53 Rev. 5 control concepts. Use when assessing or defining controls for data access, privacy, PII handling, encryption, logging, monitoring, sharing, retention, segregation of duties, or security exceptions. Use data-standards-management for enterprise standard requirements and use data-governance for operating model and decision rights.

bkjohn2018 Skill

Defines the domain objects, relationships, statuses, and lifecycle states needed for governance and finance-led operating dashboards. Use when designing dashboards for policies, controls, evidence, issues, certifications, lineage, or audit readiness.

bkjohn2018 Skill

Captures and validates data model requirements using Hoberman-style conceptual/logical framing plus Data Model Scorecard quality categories, designed to complement Kimball dimensional delivery. Use when users need better model clarity, scope control, and formal model quality scoring before implementation.

bkjohn2018 Skill

Executive Summary Writing

bkjohn2018 Skill

Governance Writing Style Guide

jdsadminllc Bundle

Dark AI/tech aesthetic with neon accents for futuristic agent-driven interfaces.

felvieira Skill

Skill para desenhar prompts reutilizaveis de texto, imagem e video com foco em clareza, controle, custo e reprodutibilidade. Use quando a qualidade do prompt for parte central da feature ou do fluxo. Trigger em: "prompt engineer", "prompt engineering", "melhorar prompt", "otimizar prompt", "template de prompt", "system prompt", "few-shot", "chain of thought", "reduzir custo de prompt", "iterar prompt", "prompt reutilizavel".

felvieira Skill

Captura screenshots via Playwright MCP de URLs/elementos pra usar em posts de blog ou documentação. Lida com viewport, cookie banners, full-page vs section, scroll a âncora, formatos PNG/JPG. Compõe com skill 41 (blog-publisher). Trigger em: "tira print", "screenshot do site", "captura tela", "screenshot da página", "print da landing", "print do dashboard", "screenshot pro blog".

felvieira Bundle

Skill especializada em paralelizar trabalho via subagents corretamente — sem cair na armadilha skill-vs-agent. Use quando precisar despachar N slices verticais, N reviews, ou N tarefas independentes em paralelo, e quando reviewers/agentes paralelos discordam sobre o mesmo achado e a decisão de arbitrar/bloquear precisa de um critério formal. Trigger em: "paralelize", "N slices", "dispatch paralelo", "worktree paralelo", "5 workers", "dispatch em paralelo", "fan-out", "scatter-gather", "comprehensive review", "multi-agent parallel", "reviewers discordam", "achados divergentes", "arbitragem entre agentes", "quem decide quando os agentes discordam".

felvieira Skill

Encontra oportunidades de "deepening" no codebase — refactors que transformam modulos shallow (interface complexa, implementacao simples) em deep (interface simples, implementacao rica). Foco em testabilidade e AI-navigability. Use quando usuario quiser melhorar arquitetura, encontrar oportunidades de refactor, consolidar modulos acoplados, ou preparar codebase para trabalho de agente. Trigger em: "deepening", "deep module", "shallow module", "refactor architecture", "improve architecture", "consolidate modules", "agent-friendly codebase", "AI-navigable", "module depth".

felvieira Bundle

Skill para desenhar e implementar integracoes de IA em aplicacoes, separando provider, adapter, hooks, observabilidade, custo e seguranca. Use quando o usuario quiser adicionar texto, imagem ou video ao app. Trigger em: "integrar IA no app", "AI integration", "provider adapter de IA", "feature de IA", "chamar OpenAI", "chamar Anthropic", "Claude API", "LLM no app", "arquitetura de IA", "custo de IA", "fallback de provider", "rate limit de IA".

dirtytrii Bundle

GStack guard method for strengthening scope boundaries, forbidden actions, stop conditions, and validation gates for an agent or workstream.

dirtytrii Bundle

Operate the WeChat Official Account AI-application workflow for articles, weekly digests, rich formatting, draft-box updates, and local handoff. Use for Agents, MCP, Skills, tool calling, workflows, and open-source AI updates; exclude Xiaohongshu and Xianyu.

dirtytrii Skill

Use when splitting content work between a raw drafting model/window, an editor-in-chief, and the user's final taste review. It keeps prompts loose enough for judgment while preserving facts, account boundaries, and publish gates.

dirtytrii Bundle

Route CEO-first multi-window work into owner-led role loops. Use for 总控/CEO、架构/CTO、内容主编、角色派发与复用、模型路由、来源窗口回调、role-windows 台账、skill 命中统计或 multi-window loop engineering。

dirtytrii Bundle

Audit and optimize a skill system with evidence. Use for skill architecture, token or context waste, role and prompt boundaries, routing hits or misfires, plugin packaging, registry/docs/source drift, duplicate workflows, or deciding whether a reusable change belongs in a skill. Runs deterministic checks first and applies only authorized, minimal fixes.

pedroiff0 Bundle

vLLM: high-throughput LLM serving, OpenAI API, quantization.

pedroiff0 Bundle

One-line description of what this skill does and WHEN to use it. Be specific — this is how the agent decides to load it.

pedroiff0 Bundle

Consume the SUAP (Sistema Unificado de Administração Pública) REST API v2 used by Brazilian federal institutes (IFRN, IFF, IFS, etc.) — obtain a JWT via /api/v2/autenticacao/token/, discover real endpoints via /api/openapi.json, and pull student data (boletim, períodos letivos, dados do aluno). Covers the WAF User-Agent gotcha, the username/password JSON body format (NOT Basic Auth), the /api/ensino/ path layout, and the scope limits of the simple token. Use whenever a user wants to script/automate access to SUAP academic data with matrícula + senha, or debug a 403/422/400 from the SUAP token endpoint.

pedroiff0 Bundle

Install and verify NVIDIA proprietary GPU drivers on Debian (including trixie/13) so local LLM tools (Ollama, llama.cpp, vLLM) can use the GPU. Covers nouveau blacklist, initramfs, the trixie nvidia-smi transitional-dummy gotcha and its official .run extraction fix, and verifying Ollama actually uses VRAM. Use whenever a user wants to enable GPU acceleration for local models, install the NVIDIA driver on Debian, or debug "nvidia-smi command not found" after a driver install.

pedroiff0 Bundle

SAM: zero-shot image segmentation via points, boxes, masks.

pedroiff0 Bundle

Personal GitHub knowledge base built from the user's starred repos (pedroiff0). Maps all 41 starred repositories into knowledge domains (free APIs, sysadmin/self-hosted, Python, AI/agents/RAG, algorithms, math/edu, astronomy/science, web/Node, Hermes ecosystem, news) and provides on-demand retrieval of the relevant repo content via raw README / GitHub API. Use as a base of knowledge when the user asks for libraries, APIs, self-hosted tools, agent patterns, or references.

pedroiff0 Bundle

W&B: log ML experiments, sweeps, model registry, dashboards.

pedroiff0 Bundle

Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.

pedroiff0 Bundle

Standardize GitHub ISSUE and PR metadata (Assignee, Reviewer, Labels, Project, Milestone, Development, Relationship) and ship a strong, well-structured code-review template. Includes ready-to-use issue forms, PR template, and a severity/rubric review model.

pedroiff0 Bundle

Live catalog of skills, plugins, and agents actually installed on THIS Hermes server (pedroiff0). Lists the 233 installed skills grouped by domain, the 18 installed plugins, and confirms there are no custom agent definitions (subagents are runtime leaf/orchestrator). Includes a generator script so the catalog can be refreshed from disk on demand.

fworks-tech Skill

Monitor and debug application errors via Sentry MCP and CLI. Use when triaging production errors or managing releases.

fworks-tech Skill

Use when reviewing PRs, checking CI status, or interacting with GitHub via MCP. Requires the GitHub MCP server to be configured in opencode.json or .agenthood/config.json.

fworks-tech Skill

Conducts multi-axis code review across correctness, readability, architecture, security, and performance. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human.

2428424081cn Bundle

大模型 Prompt 提示词 A/B 测试与版本迭代指南。提供提示词量化评估指标与防御性提示词设计。

2428424081cn Bundle

Guidelines for interacting with GitHub and request permissions from the user when commands fail due to restrictions in the agent environment.

2428424081cn Bundle

Playwright 端到端自动化测试生成指南。遵循 Page Object Model (POM) 架构与弹性 Locator 定位。

2428424081cn Bundle

大模型 Prompt 注入防御与越狱防护准则：非可信数据定界隔离、系统提示词防嗅探与输入清洗。