Skills
98,639 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Train or fine-tune language and vision models using TRL (Transformer Reinforcement Learning) or Unsloth with Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment.
Write, rewrite, edit, humanize, or post text in the user's own voice without AI tells. Use when the user asks you to write as them, polish their prose, humanize or de-slop text, draft essays, posts, emails, notes, explanations, or research writing, or produce any outward-facing text where their identity, tone, or writing preferences matter.
Use when Viktor wants to generate outreach hypotheses for an IT agency client from a website URL or text brief. Performs Phase 1 (extract services / ICP / use cases / case studies from the client's site) + Phase 2 (build a 10-hypothesis matrix of 2-3 ICP × 2-3 signals × 2-3 offers, each scored with Viktor's 5-factor weighted framework). Trigger when the user says: "генеруй гіпотези для клієнта", "побудуй матрицю гіпотез", "оціни цей агентський сайт", "що тестувати для [клієнт]", "hypothesis generation for agency", "матриця гіпотез по URL", or provides a client URL and asks what campaigns to run. NOT for one-off campaign idea generation when ICP/signals are already known — use hypothesis-builder for that.
Comprehensive Hyprland Wayland compositor configuration skill. Use when users need help with: (1) Creating or modifying Hyprland config files, (2) Setting up keybindings, window rules, monitors, or animations, (3) Troubleshooting Hyprland configuration issues, (4) Searching for valid config variables and values, (5) Understanding Hyprland syntax and structure, (6) Setting up multi-monitor configurations, (7) Configuring input devices, decorations, or layouts, or (8) Any other Hyprland-related configuration tasks.
Use when the user asks to generate images, illustrations, logos, mascots, covers, posters, OG images, favicon source images, or realistic scene images, or when a website page lacks images or still has placeholders during site building. Includes requests like “draw one,” “make a set of images,” “the website needs images,” and “image gen.” This Skill handles prompts, generation, three-stage review, compression, and saving files. Use codex when asking Codex CLI to do a coding-agent task, agent-fleet for ordinary task dispatch, and rankup for page SEO and launch gates.
Impacket Windows protocol toolkit. Use when remote operations are needed over Windows protocols such as SMB/Kerberos/WMI/DCOM/MSSQL. Covers secretsdump credential dumping, psexec/wmiexec remote execution, Kerberos attacks, and NTLM relay; it is a foundational toolkit for domain penetration and Windows lateral movement
Run the guided, end-to-end loop that makes one Duvo Agent better. Use when the user wants to "improve this Agent", "make my Agent better", "help me tune this Agent", "this Agent isn't good enough", "fix this Agent based on its Runs", or asks you to take an Agent from where it is to a concrete, applied improvement — not just a read-only audit. You drive the conversation: pin the Agent, survey its whole setup (AOP, Connections, Skills, triggers, memory, recent Runs), agree on what to base the improvement on (the user's feedback or the evidence in the Runs), propose a Run range and confirm it, get the grounded findings from the workflow-debugger / run-debugger skills, propose concrete changes, and — on the user's yes — apply them (AOP via aop-writer as a new Build, config with your tools). The boundary against workflow-debugger: that skill is the read-only audit; this skill is the interactive loop that ends in an applied change and calls workflow-debugger to do the analysis.
Run the guided, end-to-end loop that makes one Duvo Queue — and the producer→consumer workflow around it — better. Use when the user wants to "improve this Queue", "this Queue keeps backing up, fix it", or take a Queue to a concrete, applied improvement — not just a read-only audit. You drive the conversation: pin the Queue, survey its whole shape (producers and consumers, backlog and case state, both Agents' AOPs at the seam, triggers and concurrency, the Runs flowing through), agree what to base the improvement on (the user's feedback or the evidence in the Runs/backlog), propose a Run/case range and confirm it, get grounded findings from the workflow-debugger / run-debugger skills, propose concrete changes, and — on the user's yes — apply them (AOPs via aop-writer as new Builds, topology and triggers with your tools). The boundary vs workflow-debugger: that skill is the read-only audit; this is the interactive loop that ends in an applied change. For a single standalone Agent, use improve-agent instead.
Fills in missing data for inbound leads — researches the company, identifies the person's role and seniority, finds other stakeholders at the company, checks for existing CRM relationships, and updates the lead record. Produces enriched lead data ready for qualification or outreach. Tool-agnostic.
Qualifies inbound leads against full ICP criteria — company size, industry, use case fit, role/seniority of the person. Checks CRM and existing customer base for duplicates and existing relationships. Outputs a scored CSV with qualification status, reasoning, and pipeline overlap flags. Tool-agnostic — works with any CRM, enrichment tool, or data source.
Generates in-depth research reports for a specified industry based on the East Money database. This Skill should be triggered when a user's question contains an identifiable industry/sector/domain subject and the intent is industry understanding, industry analysis or writing a research report. Common phrasings include "XX industry research", "XX industry report", "help me analyze industry XX", "in-depth research on industry XX", "market analysis of domain XX", etc. Even if the user does not explicitly say "report", it also applies whenever an industry is the core subject and systematic or in-depth research/analysis is requested.
Make a single KBAC authorization decision via the IndyKite AuthZEN REST API (`POST /access/v1/evaluation`) - returns a boolean `decision` for one (subject, action, resource) triple, optionally with per-request `context.input_params`. Use for a single yes/no question - "can ada PROVISION gpu-node-7?", "is this user allowed to delete this document?", "gate this operation on a live check", or debugging why one decision is false. Not for many checks at once (use indykite-authzen-evaluations), not for enumerating which actions/resources/subjects are allowed (use indykite-authzen-search-action / -search-resource / -search-subject), and not for authoring the policy behind the decision (use indykite-authzen-kbac-policies). For the same decision over MCP/JSON-RPC see indykite-mcp-server (`authzen_evaluate`).
Run many KBAC authorization decisions in one call via the IndyKite AuthZEN REST API (`POST /access/v1/evaluations`), with top-level subject/action/resource/context as defaults overridden per entry; returns one `decision` per entry, in order. Use when checking a known, fixed set of checks at once - one subject against many resources, one action across many subjects, or any mix of triples - e.g. "of these servers, which can grace provision?", "for each of these users, can they deploy gpu-node-7?". For a single check use indykite-authzen-evaluation; to enumerate ALL allowed actions/resources/subjects (open-ended, not a fixed list) use indykite-authzen-search-action / -search-resource / -search-subject; to author the policy use indykite-authzen-kbac-policies.
List the actions a subject is allowed to perform on a resource via the IndyKite AuthZEN REST API (`POST /access/v1/search/action`) - returns the granted action names for one pinned (subject, resource) pair. Use to enumerate permitted operations - "what can linus do with gpu-node-7?", "which actions does this user have on this item?" (e.g. to render only allowed UI controls). Not for a specific-action yes/no ("can linus DEPLOY gpu-node-7?" -> indykite-authzen-evaluation); to enumerate the other axes use indykite-authzen-search-resource (which resources) or indykite-authzen-search-subject (which subjects); to author the policy use indykite-authzen-kbac-policies.
List the resources a subject is allowed to perform a given action on via the IndyKite AuthZEN REST API (`POST /access/v1/search/resource`) - given a subject and an action, returns the matching resource instances of a type. Use to enumerate permitted resources - "which servers can linus provision?", "list the documents this user can read" (access-filtered feeds). Returns `{type,id}` references, not a yes/no decision and not the resource data itself (for graph data use indykite-ciq-read). For a single-resource yes/no use indykite-authzen-evaluation; to enumerate the other axes use indykite-authzen-search-action (which actions) or indykite-authzen-search-subject (which subjects); to author the policy use indykite-authzen-kbac-policies.
List the subjects allowed to perform a given action on a resource via the IndyKite AuthZEN REST API (`POST /access/v1/search/subject`) - given a resource and an action, returns the matching subject instances of a type. Use to enumerate who has access - "who can provision gpu-node-7?", "list the people allowed to approve this document" (audit / reviewer views). Not for a specific-subject yes/no ("can grace provision gpu-node-7?" -> indykite-authzen-evaluation); to enumerate the other axes use indykite-authzen-search-action (which actions) or indykite-authzen-search-resource (which resources); to author the policy use indykite-authzen-kbac-policies.
Build modern SPAs with Inertia.js and Rails using React, Vue, or Svelte. Use when creating Inertia pages, handling forms with useForm, managing shared props, or implementing client-side routing. Triggers on Inertia.js setup, SPA development, or React/Vue/Svelte with Rails.
Build self-contained interactive HTML dashboards with Chart.js, dropdown filters, and professional styling. Use when creating dashboards, building interactive reports, or generating shareable HTML files with charts and filters that work without a server.
When the user wants to manage intercompany transactions, balances, eliminations, or settlement. Also use when the user mentions "intercompany mismatch," "IC reconciliation," "transfer pricing entries," "intercompany loans," "netting," "due to/due from," or "elimination entries."
iOS app penetration testing methodology. Covers IPA static analysis (decompilation/Plist analysis/binary inspection), dynamic analysis (Frida/Objection/Cycript), data storage security (Keychain/NSUserDefaults/CoreData), network communication security, jailbreak detection bypass, and URL Scheme abuse. Trigger when an agent needs to test iOS app security, analyze IPA files, or bypass iOS protection mechanisms.
Collect, organize, and validate evidence for ISO 27001 and SOC 2 audits. API-first approach with CLI commands for major cloud platforms. Produces timestamped, auditor-ready evidence packages. Use when user says "collect audit evidence," "prepare evidence package," "evidence for the auditor," "refresh evidence," or "evidence gap analysis."
Create a GitHub issue in the tldraw repository from a user description, then mature it through follow-up questions. Use when the user invokes issue, asks to create an issue, report a bug, file a feature request, or answers follow-up questions for an issue created by this skill.
Reviewer-gated iterative fleet for headless `claude -p`, `codex exec`, or `pi -p` workers that run in cycles until a designated reviewer approves the output. Use when the work needs multiple rounds of iteration with a quality gate — a reviewer worker reads all worker logs, writes a verdict (lgtm | iterate | escalate), and the orchestrator decides whether to continue, pause, or stop. NEVER kills or restarts workers automatically; the operator owns all kill/pause decisions.
Implement comprehensive testing strategies using Jest, Vitest, and Testing Library for unit tests, integration tests, and end-to-end testing with mocking, fixtures, and test-driven development. Use when writing JavaScript/TypeScript tests, setting up test infrastructure, or implementing TDD/BDD workflows
Write Jobs-to-be-Done (JTBD) job stories and map customer jobs across functional, social, and emotional dimensions. Use when defining user needs, writing job stories, conducting JTBD research, or reframing features around customer outcomes. Produces a job story map with opportunity scoring, pain intensity ratings, and product opportunity analysis.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.