AI & ML

vathymut Bundle

Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs the domain model maintained.

vathymut Bundle

Use when a Python import fails or you need to choose between competing libraries (plotting, serving, tabular DL) or route a package to the correct feature (dev/agent/default).

jaquevan Bundle

Scans today's Cursor and Claude agent transcripts, distills what was worked on, changed, deleted, and why, then creates a tagged daily entry in the Second Brain wiki. Use when the user says "log my sessions", "what did I do in cursor today", "save my session history", "daily session log", "log today's work", "end of day log", "journal my sessions", or asks to record what they worked on in agent conversations.

jaquevan 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.

jaquevan Skill

Looks up a Red Hat employee's title, location, and manager using the Dataverse MCP and the roverpeople data product. Use when the user says "who is", "look up", "what is their title", "who manages", "where is <person> located", "rover lookup", "find <person> in rover", or asks about someone's role, manager, or office location at Red Hat. Also use when the user asks "what team is <person> on" or "who does <person> report to".

jaquevan Bundle

Reviews a SKILL.md or prompt file against 10 structural conventions from Zack Bodnar's automated-usability-testing project and the Anthropic skill authoring guide. Checks separation of concerns, examples (good+bad), tables vs prose, output templates, anti-patterns, numbering, length, data vs instructions, confidence ratings, and optional vs required markers. Cites specific lines and suggests fixes. Use when the user says "review this skill", "check my skill", "skill review", "does this skill follow conventions", "audit this SKILL.md", "check this prompt file", or wants feedback on a skill before publishing.

jaquevan Bundle

Cross-agent task dispatch system. Send tasks to other repos, check for agent replies, scan repo activity (git commits, agent sessions), and manage bidirectional conversations between the claw and agents in other workspaces via BRAIN_TASKS.md files. Use when the user says "send a task to", "dispatch to", "add a task to", "check agent replies", "what tasks are pending", "scan my repos", "what have I been working on", "project activity", "check on my other projects", or mentions dispatching work to another repo or checking what agents have done.

jaquevan Skill

Unified morning briefing that pulls calendar, tasks, email, Slack threads, Jira sprint status, cross-repo activity, and agent replies into one structured overview with meeting classification and priority ordering. Use when the user says "daily briefing", "prep me for my day", "morning brief", "what's on my plate", "what's on deck", "start my day", "daily prep", "what do I have today", or asks for a summary of their day ahead.

jaquevan Bundle

Designs and builds agent loops, goals, skills, and automations in Cursor. Walks through loop type selection, the five-piece framework, prompt writing, and routes to the right skill for execution. Use when the user says "build a loop", "create an automation", "set up a goal", "design a skill", "make my agent prompt itself", "loop engineering", "agent loop", "design a loop", "write a goal", "scheduled task", "recurring agent", or wants to set up any recurring, scheduled, or goal-based agent workflow.

jaquevan Bundle

Manage <YOUR_JIRA_PROJECT> sprint transitions using the Atlassian MCP. Finds your assigned tickets, shows sprint and story point status, and updates tickets for the next sprint. Use when the user says "sprint planning", "move tickets to sprint", "storypoint my tickets", "sprint review", "what's in my sprint", "sprint kickoff prep", "end of sprint", "sprint transition", "break down this ticket", "story point this", or "plan my sprint".

tmforum-oda Skill

Given a TMFSxxx use-case id, processes every image in its knowledge/use-cases/{ID}/media/ folder in turn -- classifies it, renames it descriptively, reverse-engineers UML sequence diagrams and information/data-model diagrams into PlantUML .puml files, and writes a text-description.md for everything else -- updating the use case's own document body to match. Use this to enrich a use case's raw extracted images into agent-friendly formats. Unlike other skills here, this one writes to knowledge/, not just reads it.

stacksjs Skill

Use when working with fake data generation in a Stacks application — seeding databases, generating test data, model factories, or using faker utilities. Covers @stacksjs/faker (wrapper around ts-mocker) and its integration with the database seeder.

stacksjs Skill

Use when implementing search in Stacks — full-text search with Meilisearch or Algolia backends, document indexing, search settings management, the useSearch model trait for automatic indexing, or search driver configuration. Covers @stacksjs/search-engine and config/search-engine.ts.

florafauna-ai Skill

Turn a product into a shoppable PDP asset set and an annotated PDF. Drop a product photo into the chat -- and a model frame if you have one -- and it returns clean studio plates, on-model views, and an A4 deck. Works from a single image. The product is reference-locked and never re-described. Use when someone asks for PDP assets, product detail page imagery, an on-model set, a product gallery, shots of this product on a model, or a product recreated de-branded and shot on a person. Do not use for placing finished artwork in the world -- that is flora-mockup-deck -- or for a one-off image with no set around it.

florafauna-ai Skill

Entry point for FLORA's MCP server. Use this skill when (1) an agent has the FLORA MCP connected and needs to decide which flora-* skill applies to a request, (2) the request involves generating images, video, or audio, running a saved FLORA technique, or reading and writing a FLORA canvas, (3) the user has a list of products, SKUs, or variants and wants consistent assets, or (4) a first FLORA result came back wrong and needs to be driven to a goal. Routes to the specialised flora-* skills; does not generate on its own.

florafauna-ai Skill

Drive a FLORA generation to a stated goal with a generate → evaluate → refine loop that stops. Uses a free vision judge to score each candidate against the goal, names the specific defect, changes the prompt to fix it, and halts when the goal is met, the defect stops moving, or the budget cap is hit. Use when the user has a precise target and the first result is wrong — "make it more…", "that's not quite right", "closer but the X is off", "keep going until it looks like Y". Use it especially when someone is re-running the same prompt hoping for a better roll. Do not use for exploratory work with no defined goal, and do not use to generate many different items — that is flora-batch-generate.

florafauna-ai Skill

Open a specific FLORA project canvas and get oriented before doing work — sign an in-harness browser into the canvas with WebMCP when the agent has one, otherwise work through the hosted MCP tools, then read the canvas and propose what to do next. Use when the user names a workspace and project ("open my Meridian project", "let's work on this canvas"), pastes a canvas link, or arrives from FLORA's "Open in agent" button. Do not use to create a new project.

florafauna-ai Skill

Run a saved FLORA technique — a multi-step visual workflow such as background swap, relighting, upscale, model swap, or sketch-to-render. Use when the user names a technique, asks to transform an image they already have ("put this on a clean background", "match this lighting", "upscale this", "make this look photographed"), or asks what FLORA can do. Do not use for a plain text-to-image request with no source image, and do not use to inspect existing project work.

florafauna-ai Skill

Turn a product into a shoppable PDP asset set and an annotated PDF. Drop a product photo into the chat -- and a model frame if you have one -- and it returns clean studio plates, on-model views, and an A4 deck. Works from a single image. The product is reference-locked and never re-described. Use when someone asks for PDP assets, product detail page imagery, an on-model set, a product gallery, shots of this product on a model, or a product recreated de-branded and shot on a person. Do not use for placing finished artwork in the world -- that is flora-mockup-deck -- or for a one-off image with no set around it.

florafauna-ai Skill

Drive a FLORA generation to a stated goal with a generate → evaluate → refine loop that stops. Uses a free vision judge to score each candidate against the goal, names the specific defect, changes the prompt to fix it, and halts when the goal is met, the defect stops moving, or the budget cap is hit. Use when the user has a precise target and the first result is wrong — "make it more…", "that's not quite right", "closer but the X is off", "keep going until it looks like Y". Use it especially when someone is re-running the same prompt hoping for a better roll. Do not use for exploratory work with no defined goal, and do not use to generate many different items — that is flora-batch-generate.

florafauna-ai Skill

Run a saved FLORA technique — a multi-step visual workflow such as background swap, relighting, upscale, model swap, or sketch-to-render. Use when the user names a technique, asks to transform an image they already have ("put this on a clean background", "match this lighting", "upscale this", "make this look photographed"), or asks what FLORA can do. Do not use for a plain text-to-image request with no source image, and do not use to inspect existing project work.

shrimpwtf Bundle

Use for "interrogate", "adversarial review", "multi-model review", "challenge this", "stress test this code", "find blind spots", or "tear this apart". Multiple LLM reviewers challenge changes from independent angles.

shrimpwtf Bundle

poteto's agent style for concise, detailed responses, deliberate subagents, unslopped prose, simple code, and verified work. Use for poteto-mode or requests to work in this style.

shrimpwtf Skill

Configure Benny and prepare its triage and repro automations. Use when installing Benny or changing its Slack, tracker, repository, routing, control, model, or budget settings.

aze333sun Skill

Use when searching for skills, browsing skill catalogs, installing a skill, or extending capabilities with reusable AI agent components from the prompts.chat registry.

kinginyellows Skill

Validate Graphite CLI prerequisites and configure settings for AI agent workflows. Use when first installing the plugin, after Graphite auth changes, or when gt commands fail.

kinginyellows Skill

Audit checklist for evaluating an existing codebase against agent-native architecture principles to identify parity gaps. Use when auditing for agent-native readiness, before adding a new agent capability, or when triaging agent-behavior regressions.

kinginyellows Skill

Reference for agent-native architecture principles: action parity, context parity, shared workspace, primitives over workflows, and dynamic context injection. Use when authoring agent integrations or designing tool surfaces.

grandamenium Skill

Browse the community catalog for new skills, agent templates, and org templates. Discover what is available and recommend useful items to the user.

grandamenium Skill

Package a local skill, agent, or org template for community sharing. Strips PII, scans for secrets, prepares a clean submission, and opens a PR to the community catalog.

grandamenium Skill

Daily goal lifecycle management. Use for: morning briefing goal cascade, setting daily focus, refreshing agent goals, reviewing goal progress. Triggered daily as part of morning review.

grandamenium Skill

Daily git snapshots of agent workspace changes. Stages files with safety checks, reviews diff for PII, commits with descriptive message. Never pushes automatically.

grandamenium Skill

A message has just arrived in your session from the fast-checker daemon — you see a block starting with === TELEGRAM or === AGENT MESSAGE. Read it, decide what action to take, and reply using the command shown in the message header. If it is from the user, they are waiting for your response right now. If it is from another agent, they may be blocked on your reply. Handle all messages before returning to other work.

grandamenium Skill

You are about to start any meaningful piece of work — a research task, a build, a draft, a deployment, anything with a deliverable. Before you begin, create a task. During the work, mark it in_progress and write WORKING ON to daily memory. When done, complete it with a result summary and log a task_completed event. If you are blocked, set the task to blocked with the reason. If a human needs to do something first, assign it to them. Without a task, your work is invisible to the dashboard and the user.

grandamenium Skill

You are about to take an action that affects the outside world, cannot be undone, or involves real people — and you have not yet received explicit permission. This includes: sending any email or message to a real person, deploying code to production, posting on social media, making a purchase or financial commitment, deleting files or data, merging a PR to main, or publishing anything publicly. Stop, create an approval, block your task, and notify the user. Do not proceed until you receive the approval decision in your inbox.

grandamenium Skill

You have hit a blocker that is not a permission issue — it is a capability issue. You genuinely cannot complete the next step because it requires a human: making a payment, entering credentials for a service you cannot access, physical action, a decision that only the user can make, or anything else outside your capabilities. You need to create a clear [HUMAN] task with step-by-step instructions, block your own work on it, and notify the orchestrator so this surfaces in the next briefing.