AI & ML

cryptopafi Skill

Control websites through the local agent-browser CLI for navigation, screenshots, form interaction, login flows, text extraction, and website-access testing. Use when auditing whether websites are reachable or extractable, when the Codex in-app Browser plugin is unavailable/insufficient, or when a repeatable shell-driven browser workflow is needed. ANTI-PATTERN: Do not use for pure HTTP checks that curl can answer, source-code review, or restricted authenticated content without credentials.

cryptopafi Skill

Turn any task into a prompt package optimized for Claude Fable 5 / Mythos 5 — recommends effort level, selects only the behavior blocks the task needs (anti-overplanning, brevity, progress-grounding, subagents, memory, send_to_user, etc.), and enforces the Fable 5 footguns (no reasoning-extraction, Opus 4.8 refusal fallback, adaptive-thinking-only). Use when the user says /prompt-fable5, 'prompt fable 5', or 'optimize this for Fable 5'. ANTI-PATTERN: not for general prompt optimization with no Fable 5 target (use /opt), and it does not execute the built prompt.

cryptopafi Bundle

Full NexusOS Self-Optimization Loop (SOL) procedure for continuous prompt improvement: discover, audit, build, approve, apply, track, and persist prompt optimizations. Use when Pafi says /sol, asks for SOL governance, or requests the complete continuous prompt-improvement workflow.

cryptopafi Skill

Enforce read-only RUO Peptides/PeptraLabs Supabase access for non-Hermes agents and route every mutation as a validated instruction set to Hermes-main/default, the sole mutation owner. Use for RUO reads, write requests, migrations, schema/config/RLS/storage/RPC changes, syncs, or agent access. ANTI-PATTERN: Never use this skill to authorize or execute a direct non-Hermes Supabase write.

iopsystems Skill

Capture a conversation or working session as a **discussion** (multiple attendees) or **session note** (solo author, often with an agent) in the knowledge-iop vault, AND surface what existing artifacts the conversation implicates. Discussions are the vault's napkin — freely editable and deletable — but their real value is triaging what they imply. Use this whenever the user says things like "let's write this down", "record the meeting", "capture this session", "note the decision points", "what did we discuss", or when a conversation has clearly concluded a loop that should be logged. This skill also surfaces follow-up actions: does this discussion invalidate an existing brief? does it warrant a new inquiry? does it frame a problem that isn't in the vault yet?

iopsystems Skill

Record a **decision** — the "because" that cites one or more design briefs (each of which frames a problem brief). Three-hop traceability: why do we do this? → this design → because of this problem. Use this when the user says things like "record the decision", "capture what we decided", "let's log this decision", "add a decision", or after a design brief has been accepted and the user wants to capture the downstream commitment. Decisions are written by humans (with agent help); briefs are written by agents. Keep that boundary firm. A decision cannot be `accepted` while any of its `derived_from` design briefs is not accepted — this skill invokes `vault_check_transition` to enforce that before writing.

iopsystems Bundle

Use when a user asks which active skills from this MCP skill catalog or embedded agent-skill templates to use, seed, or adopt for a project; or when checking whether a project-local instance already covers that catalog need.

iopsystems Bundle

Exchange threaded review notes with another local agent through a repo-local git bridge. Use when asked to "review this branch", "get a second opinion", "check the review feedback", "reply to the review", "what did Codex say", or whenever two independent agents are reviewing each other's work on the same repository. Installs itself into a repository that does not have the bridge yet, and upgrades one whose format has drifted. Symptoms that this skill applies: a review that arrives as chat text and is lost on the next compaction, a finding you disagree with and have nowhere to dispute, a reviewer repeating a finding that was already answered, or a repository where the other agent's notes have no agreed home.

iopsystems Bundle

Use when creating or materially revising system architecture diagrams — a build-time structure chart (units, layers, product composition) or runtime charts (thread model, life-of-a-request flow). Triggers include "architecture diagram", "threading diagram", "dataflow chart", "life of a request", "diagram the system". Symptoms include a hand-maintained diagram drifting from the code, a chart that cannot say whether a thread or queue still exists, or structure and runtime claims tangled in one unreadable picture.

iopsystems Bundle

Use when creating or materially revising system architecture diagrams — a build-time structure chart (units, layers, product composition) or runtime charts (thread model, life-of-a-request flow). Triggers include "architecture diagram", "threading diagram", "dataflow chart", "life of a request", "diagram the system". Symptoms include a hand-maintained diagram drifting from the code, a chart that cannot say whether a thread or queue still exists, or structure and runtime claims tangled in one unreadable picture.

aiagentwithdhruv Skill

Answer Euron Gen AI Bootcamp doubts instantly. Covers Euri API setup across all frameworks (Python, LangChain, OpenAI SDK, TypeScript, n8n, curl), MCP server setup, common errors, model recommendations, and token limits. Use when someone in the bootcamp asks a question.

aiagentwithdhruv Bundle

Scrape and verify business leads using Apify, classify with LLM, enrich emails, and save to Google Sheets. Use when user asks to find leads, scrape businesses, generate prospect lists, or build lead databases for any industry or location.

aiagentwithdhruv Bundle

Classify leads using LLM for complex distinctions like product SaaS vs agencies. Use when user asks to classify leads, filter leads by type, or categorize businesses.

dailybothq Bundle

DeepWorkPlan — turn any repo AI-first and run Deep Work Plans. Routes to create, execute, refine, resume, status, verify, and repo-onboarding sub-skills based on intent. Use when the developer wants to plan, execute, manage, or verify structured multi-task work, or make a repository AI-agent-ready.

dailybothq Bundle

Author or update reusable skills, agents, and commands in the current repo — reason about the repo's .agents/ layout, follow the Open Agent Skills frontmatter contract, and keep the .agents/docs/ catalog in sync. Use when a developer wants to create or evolve the repo's agent kit (skills, agents, commands), or runs /skill-create or /agent-create.

yafet-dev Bundle

Advanced clickjacking and UI redress testing methodology for bug bounty and application security work. Use when testing or reviewing frame embedding, OAuth authorization clicks, donation/payment flows, admin actions, browser permission prompts, double-clickjacking, drag-and-drop abuse, UI overlay attacks, mobile WebViews, postMessage UI flows, and any sensitive action that can be triggered by deceiving a user's clicks.

alexisbalayre Skill

Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.

alexisbalayre Bundle

Build a throwaway prototype to answer a design question. Use when the user wants to prototype, sanity-check whether a state model or business logic feels right, or explore what a UI should look like.

alexisbalayre Skill

Plan a huge chunk of work (more than one agent session can hold) as a shared map of decision tickets on your issue tracker, and resolve them one at a time until the way to the destination is clear.

alexisbalayre Skill

Cost-optimal multi-agent code review of local changes or a PR, across six relevance-gated, model-tiered areas, with a deterministic pre-flight and a record-all structured verdict CI renders to the PR in full.

alexisbalayre Bundle

Build and sharpen a project's domain model. Use when discussing codebase terminology, writing or editing `docs/glossary.md`, or recording or editing an ADR.

shipshitdev Bundle

Creates MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Activates on: "build an MCP server", "create MCP tools", "integrate API via MCP", "write a FastMCP server", "add an MCP server to an agent", or any request to wrap an external API as LLM-callable tools in Python or Node/TypeScript.

shipshitdev Bundle

Expert guide on prompt engineering patterns, best practices, and optimization techniques. Use when user wants to improve prompts, learn prompting strategies, debug agent behavior, or design content generation prompts.

ralf-claude-skills Skill

Run any question, idea, or decision through a council of 5 AI advisors who independently analyze it, peer-review each other anonymously, and synthesize a final verdict. Based on Karpathy's LLM Council methodology. MANDATORY TRIGGERS: 'council this', 'run the council', 'war room this', 'pressure-test this', 'stress-test this', 'debate this'. STRONG TRIGGERS (use when combined with a real decision or tradeoff): 'should I X or Y', 'which option', 'what would you do', 'is this the right move', 'validate this', 'get multiple perspectives', 'I can't decide', 'I'm torn between'. Do NOT trigger on simple yes/no questions, factual lookups, or casual 'should I' without a meaningful tradeoff (e.g. 'should I use markdown' is not a council question). DO trigger when the user presents a genuine decision with stakes, multiple options, and context that suggests they want it pressure-tested from multiple angles.

ralf-claude-skills Skill

Use this skill to BUILD, create, edit, debug, or extend n8n workflows via the N8N MCP connector -- especially when the user pastes course content, a lesson/unit, exercise, assessment brief, or step-by-step spec expecting a working workflow in their n8n instance. Trigger on requests to build, implement, or fix a unit, exercise, workflow, sub-workflow, AI agent workflow, data table, or webhook workflow, or on pasted content resembling an n8n lesson (mentions HTTP Request, IF, Merge, Switch, Set/Edit Fields, Data Table, Aggregate, Summarize, Convert to File, Loop Over Items, Webhook, Respond to Webhook, Error Trigger, Execute Sub-workflow, Chat Trigger, AI Agent, assessment ID, X-Assessment-ID). Also trigger on follow-ups like "next unit" or "build this too". Covers: paste-confirm-build loop, required question format, N8N SDK build pattern (validate -> create/update -> report), node config recipes, credential/assessment-ID handling, gotchas (wrong auto-assigned credentials, publish requirements, pinned data).

ralf-claude-skills Skill

Creates a clean empty Notion template skeleton from any idea — databases, properties, and views, no sample data. Trigger when user says: "build the template", "create the Notion page", "make the skeleton", "set up Notion structure", or provides any template idea to build. Works for any niche or audience. Skill 4 furnishes it later. Requires Notion MCP connected.

ralf-claude-skills Skill

Reads a Notion template skeleton built by Skill 3 and enriches it into a ready-to-sell product. Trigger when user says: "furnish the template", "make it sellable", "add sample data", "enrich the notion page", "run Skill 4", or "make it ready to sell". Always runs AFTER Skill 3. Reads the live Notion page, adds sample rows, guided onboarding, callouts, seller copy, and a checklist. Works for any niche. Requires Notion MCP connected.

ralf-claude-skills Bundle

Monitors task execution for skill improvement opportunities. Use this skill during ANY multi-step task, agentic workflow, or substantive work session where the agent is using tools and producing deliverables. It captures patterns, user corrections, workflow insights, and methodology worth preserving as reusable skills. Also triggers during post-task feedback discussions and when the user explicitly mentions skill observations, improvements, the observation log, skill taxonomy, or asks the agent to watch for skill opportunities. Also known as "One Skill to Rule Them All" — trigger on this phrase too. IMPORTANT: this skill should be invoked at the start of every task-oriented session — if you are about to use tools to produce deliverables, invoke this skill first. For reliable activation, pair this description with a CLAUDE.md instruction or harness-level session-start hook (see Recommended Activation Setup) — description-level matching alone is not enforceable.

dyoshikawa Skill

Run the review-pr skill on another model (github-copilot/gpt-5.4) via opencode, then thoroughly investigate and vet the resulting findings, and present only the valid ones to the user.

autonomous-ai Bundle

Manage the device's face recognition roster — enroll new faces (3 paths: user-supplied photo, agent-captured snapshot on user request, or HAL's familiar-stranger prompt) and maintain the enrolled set (status / remove / reset). All enrolled persons are friends; strangers stay unnamed until promoted via one of the enroll flows.

autonomous-ai Bundle

Delegate tasks to coding and research agents on the computer paired through Harness, choosing a suitable existing agent when the user does not name one. Use this before computer-use, agent-management, or Autonomous Buddy when the user asks an agent on their Mac to work, including browser research. Continue tasks, inspect progress, and answer agent questions. Direct desktop control and ordinary questions to the main assistant are outside this skill; explicit Buddy requests belong to Buddy.

autonomous-ai Bundle

Open websites and apps and complete tasks on the user's paired Mac through Autonomous Buddy. Use for direct visible UI tasks such as "open Airbnb", "mở Chrome", "ghi vào Notes", forms, screenshots, and file organization. Do not use when the user asks a coding or research agent on the Mac to do the work; use harness-use instead, even if that agent will use a browser. The agent runs on the headless device; visible website/app interaction targets the paired Mac, not a browser installed on the device. Pure information research and physical device hardware use their own skills.

autonomous-ai Bundle

Legacy Autonomous Buddy control for explicitly requested Buddy coding sessions. When the user asks a coding or research agent on their paired Harness computer to work, use harness-use instead. This manages explicit Buddy desktop CLI sessions; clicking apps and screenshots use computer-use.

everyinc Bundle

Execute a plan or concrete work prompt end-to-end. Use when implementing from a plan document, a spec path, or a clear build request; use ce-debug for open-ended bugs. Use when an outer orchestrator needs implementation and local verification only, without the shipping tail.

everyinc Bundle

Create a session handoff for another agent, or resume, find, and read any user-selected continuity source. Use when work or conversation must continue without access to the current session history.

seguelacedric Skill

Generate a week of on-brand content for an ACA org. Use when the user says "generate this week's content", "I need [N] LinkedIn posts about [topic]", "write a content batch", "fill the content queue", "make ad creatives for [product]", or asks for multi-piece content tied to a brand voice. Uses ACA blueprints and publishing tools through one native ACA MCP connection.