agentik-os

@agentik-os source repo

227 skills
0 followers
17 hours ago last updated

▌
Blueprint Os · agentik-os bundle
Compile software, AI, platform, service, marketplace, mobile, web, or internal-tool ideas and existing project context into a complete, coherent, traceable Product + Technical Definition Pack before implementation planning or coding. Trigger on /blueprint, Blueprint {OS}, product blueprint, product-definition audit, recovery, revision, extension, delta, or preparation for Stepper {OS}. Also trigger on French requests such as "compile ce blueprint", "definition produit et technique", "audit de definition produit", or "prepare le blueprint pour le stepper". Preserve project decisions and stable IDs, separate evidence from assumptions, define product/UX/domain/data/API/AI/security/operations/test contracts, run gates, and continue across outputs without declaring partial work complete.

0
installs
▌
Execution Os · agentik-os bundle
Operate an LLM-first personal execution system that converts ambitions, goals, obligations, and ideas into focused commitments, protected work, shipped evidence, reviews, and adaptive recovery. Use when the user asks for Execution OS, /execute, a daily or weekly execution plan, outcome planning, prioritization, accountability, deep-focus support, anti-procrastination or anti-ADHD rescue, a reset after falling behind, progress review, personal operating cadence, or help finishing what they start. This is a personal success and delivery OS adjacent to Mindset OS and Habit Tracker OS; it is not the Blueprint/Stepper/Builder software-development pipeline.

0
installs
▌
Cookbook · agentik-os
Anthropic's OWN reference implementations, on tap. 94 recipes from anthropics/claude-cookbooks (MIT, pinned) covering RAG and retrieval, agent patterns (prompt chaining, routing, parallelization, orchestrator-workers, evaluator-optimizer), evals and LLM-judge harnesses, tool use and structured output, prompt caching and cost optimization, multimodal and PDF extraction, citations, extended thinking, the Claude Agent SDK, and Claude Managed Agents. Use BEFORE hand-rolling any of those: read the reference, adapt the code, do not re-derive it from memory. Triggers on "/cookbook", "/omg-cookbook", "anthropic cookbook", "claude cookbook", "reference implementation", "how does Anthropic do X", "is there a recipe for", or in French "recette anthropic", "implementation de reference", "comment Anthropic fait X". NOT for Claude Code harness questions (that is claude-code-guide) and NOT for the Claude API reference itself (that is the claude-api skill): this is worked EXAMPLES.

0
installs
▌
Pixelrag · agentik-os bundle
Give the agent EYES on a web page or document: render it to screenshot tiles with `pixelshot` and READ the image, instead of fetching HTML and parsing it. Charts, diagrams, infoboxes, tables and layout survive; markup noise, cookie walls and JS-rendered text stop mattering. Also queries a public 26M-vector visual index of 8.28M Wikipedia pages (no API key) and returns the matching screenshot CROP. Use when the user says "/pixelrag", "/omg-pixelrag", "screenshot this page", "look at this page", "read this page as an image", "what does this page look like", "visual search", or in French "regarde cette page", "fais une capture", "lis cette page comme une image", "recherche visuelle". NOT for scripted E2E of our own apps (Playwright, R-TEST) and NOT for open-ended agentic browsing of an unknown UI (browser-use, R-BROWSER) — this one is READ-ONLY LOOKING, one page at a time.

0
installs
▌
Powerups · agentik-os
Router into the Power-Ups corpus — 907 purchased business/marketing Claude skills (30 Power-Up packs of 406 skills + the 501-skill Ultimate Bundle) plus 272 companion assets (prompt libraries, swipe files, templates, tutorials) installed at ~/.omega/powerups. Use whenever a request matches a business, marketing, content, sales, launch, email, SEO, social, ads, pricing, course, community, podcast, newsletter, YouTube, LinkedIn, Etsy, outreach, branding, ops, HR, legal, finance, nonprofit or client-consulting task and no first-class OmegaOS skill already owns it. Also use when the user says "/powerups", "power up", "power-ups", "the skills I bought", "the 501 bundle", "youraipowerup", or in French "les skills que j'ai achetés", "le pack de 501 skills", "mes power-ups".

0
installs
▌
Ramflush · agentik-os bundle
Purge la RAM et fait le point perf du VPS — `sync` + drop des caches noyau (pagecache/dentries/inodes), compaction mémoire, puis rapport avant/après + top consommateurs RAM/CPU + load average. À lancer quand le VPS est sous charge / rame. Sans risque (les caches sont reconstruits à la demande) ; nécessite root (drop_caches).

0
installs
▌
Brainstorm Os · agentik-os bundle
Run a rigorous multi-agent imagination, evolution, and decision council with independent chambers, Founder DNA, frame fission, idea genomes, collisions, counterfactual worlds, structural mutation, incubation, adaptive specialists, adversarial debate, evidence routing, quality gates, concept lineage, experiments, and handoffs. Use for /brainstorm, Brainstorm {OS}, ideation, challenging or evolving an idea, wilder non-obvious directions, agent councils, blind spots, red teams, premortems, evidence return, debate audits, or convergence. Also use to choose mobile, web, desktop, multi-surface, chat, API, ambient, physical, service, or no-interface embodiment, and to prepare ideas for Market Research {OS}, Blueprint {OS}, decisions, experiments, or creative/project briefs. Preserve project boundaries and prior decisions, separate evidence from hypotheses, require genuine dissent, and maintain lineage across “challenge”, “continue”, “evolve”, and “go deeper”. Also trigger on French requests such as “brainstorm sur c

0
installs
▌
Reel Lint · agentik-os bundle
Note un script de reel /100 contre les patterns viraux mesurés (hook, rétention, format, CTA, outil branché) AVANT de tourner. Use when the user wants to check, score, or validate a reel script or short-form video script.

0
installs
▌
Habit Tracker Os · agentik-os bundle
Operate a conversation-first, LLM-assisted habit system for creating good habits, reducing unwanted habits, running daily check-ins, handling urges and lapses, producing adaptive reviews and visual progress reports, and integrating habits with Mindset {OS}, Life {OS}, or Omega {OS}. Use for habit setup, routine design, accountability, morning/evening check-ins, “I did it/I missed/I’m tempted” messages, weekly or monthly reviews, recovery seasons, behavior experiments, habit analytics, and durable tracking from natural-language conversations.

0
installs
▌
Storyteller Os · agentik-os bundle
Coach, mine, verify, shape, write, perform, adapt, score, and preserve truthful stories without erasing the user's voice. Use for personal stories, founder narratives, brand and customer stories, lessons, keynotes, pitches, reels, carousels, threads, podcasts, YouTube scripts, case studies, data stories, story banks, story interviews, narrative strategy, or requests such as “help me find my story,” “coach me, don't write for me,” “turn this moment into content,” “make this story deeper,” and commands beginning with /story, /mine, /interview, /deepen, /shape, /write, /adapt, /truthcheck, /score, /rehearse, or /storybank.

0
installs
▌
Acceptance · agentik-os bundle
OmegaOS autonomous browser-acceptance + self-heal gate. The terminal phase of a build: Playwright-sweeps EVERY route (200 + render), captures EVERY console error and failed network request, and walks the authenticated golden path with a real persisted write — then AUTONOMOUSLY fixes whatever it finds (missing route, dead auth bridge, console/network error, broken flow) and re-runs, looping until the sweep is fully green or a hard external blocker (a missing secret) is hit. "It builds" is never "it works" — this proves it works. Use when user says "/omg-acceptance", "acceptance gate", "test everything", "verify the app works", "browser e2e", or as the last step of /omg-new-project + /omg-planner builds.

0
installs
▌
Alexandria · agentik-os bundle
ALEXANDRIA OS — your personal librarian, knowledge architect, learning coach and intellectual sparring partner. Turns any book, idea, decision or problem into deep understanding, durable memory and concrete action, adapted to how YOU personally learn and remember. Seven functions (librarian, cartographer, teacher, learning architect, skeptic, strategist, personal archivist) and 20+ modes: X-ray a book chapter by chapter, distill an idea across many books, put authors in combat, challenge your plan with a 10-round sparring, run a council of perspectives, build flashcards and spaced-repetition, forge memory with stories and mind-palaces. Starts with /setup to calibrate on your learning style, ADHD/attention profile, memory sensibilities, domains and language. Use when the user says '/alexandria', 'librarian', 'analyse this book', 'x-ray this book', 'help me learn/remember X', 'teach me', 'challenge my idea', 'compare these books', or in French 'décortique ce livre', 'aide-moi à retenir', 'apprends-moi', 'challe

0
installs
▌
Cold Email · agentik-os bundle
Write B2B cold outbound emails and multi-touch follow-up sequences that sound human and get replies — subject lines, openers, body copy, CTAs, personalization. Use when the user says "/omg-cold-email", "/cold-email", "cold outreach", "prospecting email", "outbound", "SDR email", "follow-up sequence", or "nobody replies to my emails". R-MARKETING: outbound acquisition.

0
installs
▌
Journal Os · agentik-os bundle
Run MIRROR, a nightly daily-journal agent that reconstructs the day truthfully and turns it into evidence, spanning a one-question-at-a-time voice-first interview in French or English, strict separation of fact from emotion from interpretation from lesson, contradiction detection between declared intention and observed behavior (pattern-based, never from a single day), identity evidence in behavioral language, memory extraction, unfinished loops, a Tomorrow Protocol capped at three missions with success conditions, an identity challenge, a content handoff that exposes raw material without ever writing posts, and a reflection layer drawn from Stoicism, Jim Rohn and Taoism applied at most once per entry and always bound to that day's evidence. Use for the daily journal, evening review, day debrief, telling the agent about your day, weekly rollups, working out where the time actually went, checking whether behavior matches stated priorities, and capturing what happened before it is forgotten. Life first and cont

0
installs
▌
Ad Creative · agentik-os bundle
Generate and iterate paid-ad creative at scale — headlines, descriptions, primary text, RSA variants — for any ad platform, iterating on real performance data. Use when the user says "/omg-ad-creative", "/ad-creative", "ad copy", "generate headlines", "ad variations", "creative testing", or "Facebook/Google/LinkedIn ad copy". R-MARKETING: paid creative; reference docs document optional user-invoked third-party generation APIs (user keys only).

0
installs
▌
Browser Use · agentik-os bundle
Run an agentic browser task from natural language on the Browser Use cloud — the LLM-driven agent navigates an unknown site, extracts data, or fills a form on its own (NOT a local scripted browser). Use when the user says "/omg-browser-use", "/browser-use", "agentic browser task", "browse this site", "extract from a website", "fill out this form", "agentic web automation", or "navigate an unknown site". Boundary R-BROWSER: agentic/unknown-UI/open-ended only; deterministic scripted steps for OUR apps stay in Playwright/acceptance. Requires the paid Browser Use cloud plan + BROWSER_USE_API_KEY.

0
installs
▌
LLM Council · agentik-os bundle
Convene a multi-model Claude COUNCIL for a high-stakes, ambiguous, or irreversible decision — or whenever you want a genuine second opinion. Four different Claude models (Opus 5, Sonnet 4.6, Haiku 4.5, Fable 5) answer the SAME question independently and in parallel, then peer-review each other ANONYMOUSLY (answers shown as Response A/B/C, model identity hidden, so no model is favored by name), and an Opus 5 president synthesizes the final verdict WHILE surfacing the dissent. Inspired by Karpathy's LLM Council, re-implemented on the OmegaOS Workflow primitive. 100% Claude Code-native — NO API KEYS, no OpenRouter, no external provider, no extra cost: the members run as in-process Workflow sub-agents through your EXISTING Claude Code session. Use when the user says "llm council", "convene the council", "ask the council", "@council", "/llm-council", "second opinion", "deliberate", "multiple models", or faces a "high-stakes decision" / architecture choice / contradictory verification verdicts / a pre-commit design

0
installs
▌
New Project · agentik-os
OmegaOS end-to-end new-project pipeline — guiding (client/works) → stack pick → AUTO-PROVISION external services (Vercel/Convex/GitHub/Clerk/Stripe) from ~/.omega/provisioning → scaffold the full stack → wire every key → register the project → vision/PRD → /omg-planner (typed DAG plan) → omega plan-run (the engine executes it with can't-skip Gate + Guardian verify). Launchable from the TUI [N] New Project menu or by typing /omg-new-project (alias /omega-new-project).

0
installs
▌
Open Design · agentik-os bundle
Drive Open Design (self-hosted nexu-io/open-design) to build design systems and generate brand visuals from OmegaOS. It turns an agent into a design engine: prototypes, dashboards, landing pages, decks, images and video, rendered from a brand contract (DESIGN.md) using 150+ built-in design systems (apple, airbnb, ant, agentic, …), 100+ functional skills and render templates. Results are viewable in the browser over Tailscale (the OmegaOS server surface), so the operator SEES the build, not just files. Use for: building or applying a design system, a brand kit, a UI prototype, a pitch deck, marketing creatives for the marketing machine, product screens, or any "make it look designed" artifact that should be viewable at a link. Triggers (EN): "design system", "brand kit", "build a prototype", "landing page design", "pitch deck", "generate a dashboard", "marketing creative", "open design", "make a deck/prototype I can view". Triggers (FR): "design system", "kit de marque", "construire un prototype", "maquette",

0
installs
▌
Reel Script · agentik-os bundle
Écrit un script de reel Instagram calibré sur les patterns viraux — hook chiffré dès la 1ʳᵉ seconde, structure en 3 étapes, cliffhanger de rétention, 175 mots max (≈ 60 s). Use when the user asks for a reel script, video script, or short-form content script.

0
installs
▌
Health Energy Os · agentik-os bundle
Build and protect physical and cognitive capacity through sleep, movement, training, nutrition, recovery, stress regulation, environment design and appropriate professional care. Upstream capacity provider for Habit, Execution and Strategy OS. Contains 12 specialist agents, 18 skills, 8 protocols and 6 schemas. Use for sleep and recovery planning, training and nutrition design, stress regulation, energy audits, or deciding when to escalate to a professional. Trigger words: health, energy, sleep, movement, training, nutrition, recovery, stress, capacity; FR: sante, energie, sommeil, mouvement, entrainement, nutrition, recuperation, stress, capacite.

0
installs
▌
Intuitive Os · agentik-os bundle
Train, test and calibrate intuition as a measurable skill, spanning raw intuition capture before analytical contamination, five-layer separation (observation, interpretation, affect, intuition, decision), intuition provenance (expert pattern, weak signal, emotion, incentive signal, bias, noise), falsifiable probabilistic forecasting with resolution criteria, base rates and Bayesian updating, decision journaling and premortems, red-teaming a hunch, Brier scoring and calibration buckets by domain, pattern-card and mental-model libraries, and a 90-day observe/calibrate/compress curriculum. Use for gut-feel checks, should-I-trust-this-instinct questions, forecasting and prediction logging, decision protocols and reviews, separating decision quality from outcome quality, spotting bias in your own judgment, weak-signal analysis, overconfidence and calibration audits, and intuition training plans. Trigger words: intuition, gut feeling, instinct, hunch, calibration, forecast, prediction, Brier, base rate, decision jo

0
installs
▌
Linear Setup · agentik-os
Complete Linear feedback system setup wizard (v2.1). Installs widget with priority selector + page-aware Scope labels, API routes with hardcoded label IDs and default CTO assignee, Linear label GROUPS (FEEDBACK / SOURCE / optional AGENTS), public feedback page, and MCP integration. Auto-detects dashboard pages from sidebar to generate Scope labels. Use when user says "/linear-setup", "setup linear", "add feedback system", or "install Linear widget". One-time setup. After setup, use /linear-fix to resolve tickets.

0
installs
▌
Project Tidy · agentik-os bundle
Range un projet pollué par le sprawl des agents IA — regroupe la doc humaine dans docs/ et le tracking/junk agent visible dans agentic/, réécrit CLAUDE.md/AGENTS.md pour définir où va chaque fichier (les futurs agents s'y conforment), propose la suppression du périmé et signale les incohérences doc↔app. Ne touche JAMAIS au code, aux dotfolders système (.git/.planner/.audit/ .oracles…), ni au canon (vision/PRD/feature/step).

0
installs
▌
Seductive Os · agentik-os bundle
Build personal magnetism as a real capability, spanning presence and embodiment, conversational craft, warmth, humour, curiosity, attraction fundamentals, social calibration, reading and respecting another person's actual interest, flirtation and consent-gated escalation, dating strategy, apps and texting, style and self-presentation, rejection resilience, approach anxiety, long-term desire, and the inner game underneath all of it (self-worth, outcome dependence, authenticity). Consent-first and script-free: it builds a more attractive person, it never teaches manipulation, pressure, deception or routines that treat a person as a target. Use for charisma coaching, presence work, conversation practice, dating and romantic confidence, social calibration, style audits, app profile and texting review, post-rejection recovery, approach-anxiety ladders, keeping desire alive in a long relationship, and personal magnetism audits. Trigger words: seduction, charisma, magnetism, presence, flirting, dating, attraction, c

0
installs
▌
Context Memory Os · agentik-os bundle
Maintain one trustworthy, inspectable and permissioned memory layer so Omega and its OSs recover context without mixing facts, inferences, temporary states, projects or identities. Omega Core canonical shared context layer for every other OS. Contains 14 specialist agents, 20 skills, 7 protocols and 8 schemas. Use for memory design, context recall, fact-versus-inference separation, cross-project isolation checks, or permissioned knowledge retrieval. Trigger words: context, memory, recall, knowledge layer, fact versus inference, permissioned memory; FR: contexte, memoire, rappel, couche de connaissance, faits versus inferences, memoire permissionnee.

0
installs
▌
Wealth Capital Os · agentik-os bundle
Operate a conversational financial brain for personal cash flow, savings, emergency resilience, debt, goals, investment policy, risk and life-aligned capital allocation. Contains 12 specialist agents, 20 skills, 7 protocols and 7 schemas. Use for personal budgeting, emergency fund planning, debt payoff strategy, investment policy statements, risk tolerance, or goal-based capital allocation. Trigger words: wealth, capital, personal cash flow, savings, debt, investment policy, risk, capital allocation; FR: patrimoine, capital, tresorerie personnelle, epargne, dette, politique d'investissement, risque, allocation de capital.

0
installs

227 published skills · page 3 of 3