AI & ML

iberi22 Bundle

Configurar MiniMax M2.7 con MCP tools (web_search, understand_image) y Groq Whisper para audio. Ideal para agentes que necesitan búsqueda web, análisis de imágenes y transcripción de audio.

iberi22 Bundle

Generador de bundles de preguntas ICFES Colombia (Matemáticas, Lectura Crítica, Ciencias, Sociales, Inglés) para grados 6, 9 y 11 usando MiniMax MCP.

iberi22 Bundle

SKILL.md — Xavier Agent Onboarding

spences10 Skill

Secret-safe agent sessions via nopeek CLI. Use when user asks about secrets, .env loading, credential safety, or keeping API keys out of conversation context.

spences10 Bundle

Capture reusable session learnings. Use when updating agent instructions, adding lessons learned, preserving project conventions, or fixing repeated mistakes.

spences10 Bundle

Coordinate parallel agent work. Use when decomposing complex tasks, running review/implementation in parallel, managing task lists, or choosing shared vs isolated workspaces.

spences10 Bundle

Design and create Agent Skills using progressive disclosure principles. Use when building new skills, planning skill architecture, or writing skill content.

spences10 Skill

Navigating LLM tooling workflows. Use when choosing recall, search, SQLite, MCP, secret-safety, or skill validation tools.

spences10 Bundle

Find AI-readiness architecture issues and module-deepening refactors. Use when improving codebase architecture, testability, agent navigation, seams, or refactor tickets.

vostrikovva Skill

MongoDB with TypeScript document types and indexes. Use when the data model is document-oriented, not when the project already standardized on Postgres plus Drizzle.

future3ooo Bundle

Triage issues through a state machine driven by triage roles. Use when user wants to create an issue, triage issues, review incoming bugs or feature requests, prepare issues for an AFK agent, or manage issue workflow.

future3ooo Bundle

Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain the domain model.

future3ooo Bundle

Grilling session that challenges a plan against the existing domain model and updates CONTEXT.md/ADRs inline as decisions crystallise.

shawnxxxqazwsx Bundle

Coordinate complex multi-model development work when independent evidence, implementation, verification, or review packages can improve the outcome. Use for cross-module work, shared-resource risk, uncertain diagnosis, or an explicit request to split, self-check, or coordinate parallel work. Do not use for simple local changes or explanations.

shawnxxxqazwsx Bundle

Clarify what the user wants through a focused conversation, then synthesize the settled requirements into an accurate, self-contained instruction for an AI. Use when the user asks to work out their requirements, refine or audit a vague idea or prompt, be interviewed about a task, or summarize a discussion into a final instruction across planning, creative, research, administrative, and technical work. Do not trigger merely because an ordinary request could be more detailed, and do not execute the resulting instruction unless the user also asks for execution.

shawnxxxqazwsx Bundle

Automatically review Codex task history when the user says a skill, workflow, verification step, or agent failed to activate or behaved incorrectly; when the user re-reports a problem that the active workflow previously claimed fixed or verified; when an active user-owned workflow reaches an evidence- backed learning gate; or when the user asks to audit or improve skills. Also recover durable lessons after visible compaction. Update the narrowest user- owned skill, or create one only when no existing skill fits. Do not trigger for a first ordinary bug, a one-off correction, task length alone, or a mere milestone, and never modify system- or plugin-owned skills.

shawnxxxqazwsx Bundle

Plan, coordinate, execute, audit, package, and publish neural or hybrid training for the captured Manley Variable Mu data. Use when preparing a training-ready batch, selecting a causal model and losses, coordinating multiple machines through GitHub claims, monitoring for new capture batches, validating each usable model with Original/Target/Current audio-render forensics, or sealing a completed model/data/annotation bundle. Do not use to grant dataset eligibility, to mix primary and comparison-only targets, or to claim hardware equivalence without independent withheld evidence.

shawnxxxqazwsx Bundle

Research, fit, and validate Pigora hardware-EQ magnitude responses against frozen public evidence or owner-authorized inference targets. Use for target extraction/completion, coefficient or control-law fitting, and response acceptance. Keeps public-fit, inference engineering, and product readiness separate. Do not use for GUI-only work or private-capture training.

shawnxxxqazwsx Bundle

Plan, run, audit, and package calibrated input-output captures of physical analog compressors, limiters, and levelers for measured-response, gray-box, or neural plug-in modeling. Use for converter calibration, stimulus and parameter design, long-memory and stereo-link identification, dataset QA, repeatability, and model handoff. Do not use for ordinary compressor mixing advice, software-only preset matching, or DSP implementation after an accepted capture set already exists.

wuyue12-tech Bundle

MUST USE for internet research, search, lookup, URLs, GitHub code search, social platforms, YouTube, RSS, LinkedIn/jobs, podcasts, or Xueqiu market content. Routes 15 platforms through available CLIs, APIs, and OpenCLI; run `agent-reach doctor --json` to select an active backend. Not for processing content already supplied by the user, web write actions, or a platform already owned by a more specific skill. Read category references only as needed.

byronwall Bundle

Diagnose recent local Codex or agent-thread token use by querying llmly's usage explorer, grouping results by raw tool and session, and explaining token hotspots, tool overhead, and practical process changes. Use when asked why a recent task used many tokens, to review the last hour or day of agent work, or to compare sessions without exposing full prompts or tool results.

byronwall Bundle

Search and synthesize Codex task/thread history to recover meaning, recurring goals, effectiveness, friction, repeated user steering, common errors, reusable commands or one-off scripts, missing tools, workflow gaps, and practical next steps. Use when the user asks to review, audit, mine, compare, or learn from recent Codex tasks in the current project, named projects, a date range, or across projects; asks what keeps going wrong or what they repeatedly request; wants opportunities for scripts, skills, tooling, documentation, or process improvements; or wants an evidence-based retrospective of agent work.

byronwall Bundle

Review application data structures, fixtures, schemas, TypeScript types, persistence models, API contracts, and UI projections. Use when deciding whether current shapes are clear, durable, database-ready, easy to extend, or burdened by duplicate types and transformations. Produces direct but motivating improvements. Do not use for query-performance tuning or database operations unless data-model design is also in scope.

byronwall Bundle

Create copy-ready, repository-grounded prompts for ChatGPT Pro to review, research, evaluate, or plan work in a software repository. Use when a user wants to hand a GitHub repository to ChatGPT Pro, Deep Research, or ChatGPT agent. Do not use when the user wants Codex to do the repository work directly.

xakki Bundle

Map cheap, standard, and judgment work to resolved model slugs via AI_MODEL_*. Use to select a subagent tier, override defaults, or map mechanical, product-logic, and architecture work to models such as Haiku, Sonnet, and Opus.

xakki Bundle

Set up or audit Claude Code configuration in a repository, including CLAUDE.md, subagents, skills, MCP, settings, hooks, and Make targets.

xakki Bundle

Test that an agent's configuration actually behaves as written — run hook scripts against block/allow cases and lint settings.json permission rules and hook registration. Use after editing a guard hook, permission allow/ask/deny list, hook matcher, or MCP enable list, and when a hook silently never fires or a guard blocks a harmless command.

nirjhar17 Bundle

Extracts Slack XOXC and XOXD tokens by opening a browser for the user to log in, then configures the Slack MCP server for both Claude Code and Cursor. Use when the user says "Slack login", "extract Slack tokens", "set up Slack MCP", "configure Slack", "refresh Slack tokens", or wants to connect their Slack workspace to the AI assistant.

nirjhar17 Bundle

Show Cursor agent token usage, estimated costs, and activity patterns across all sessions on this machine. Supports summary, by-project, by-month, by-session, daily-activity, and tool-usage views. Use when asked about spending, costs, tokens, daily activity, usage patterns, or missing days.

dawnmoon1542 Bundle

提取指定时间范围内所有 pi agent session 记录，按项目分类整理为结构化 markdown 文档，保存到当前工作目录。用于定期回顾、进度汇总、工作记录归档。

dawnmoon1542 Bundle

Robust URL-to-Markdown extraction for AI agent workflows. Use when the user wants to "extract/summarize/convert a webpage to markdown" (especially WeChat mp.weixin.qq.com) and web_fetch/browser is blocked or messy. Uses a cheap probe via web_fetch first, then falls back to the official MinerU API (via the local mineru-extract skill) and returns a traceable result contract with source links.

itaywol Skill

Write a good, portable adept agent (subagent): trigger-shaped description, one job per agent, generator/evaluator separation, explicit boundaries, restricted tools. Apply when creating or editing an agent file or running `adept agent add`.

majiayu000 Bundle

Use when the user explicitly asks for $threads, Codex-native subagents, 开几个子 agent, or a GitHub issue/PR queue needing parallel lanes, worktrees, review/merge gates, and closure audit. Do not use for OS/language threads, chat/email/forum threads, or Assistants product threads unless Codex workflow orchestration is explicit.

zad111ak-ai Bundle

Configure Hermes Agent gateway platforms in DPI-restricted or censored networks — Telegram, Discord, and AI providers behind DPI/packet-inspection filters.

cwijayasundara Skill

Verify agent-framework skill packs declared in project-manifest.json#framework_skill_packs and print normal-terminal install commands for missing packs. Use after /scaffold reports PENDING MANUAL INSTALL, after manually adding a pack to the manifest, or to verify all configured packs are present.

vathymut Skill

Use when the user wants a topic investigated, docs or API facts gathered, or reading legwork delegated to a background agent.