Tap

Plans SEO strategy briefs: keyword research, topic clusters, link campaigns, GEO/AI-citation, free-tool concepts, and programmatic SEO.

Audits sites across structural, on-page, and link-profile surfaces; returns prioritized, evidence-cited findings with fix direction, saved to disk.

Reads, edits, and validates octomind config.toml with minimal changes: roles, MCP servers, limits, supervisor, compression, layers, and commands.

Creates and maintains tap registry artifacts: agent manifests, capabilities, dep scripts, and skills, lint-clean and load-resolvable.

Answers Octomind CLI, config, runtime, and supervisor questions from the owning skills, and routes build work to the right specialist agent.

Designs, dry-runs, and tunes octomind workflow pipelines: sequential, parallel, loop, and conditional steps that chain octomind run invocations.

Career coach for direction and job search: Designing Your Life, hidden-job-market strategy, ATS-aware materials, STAR story banks, and negotiation prep.

Prescribes one named training program with progressive overload, 5/3/1, volume landmarks, deloads, and C25K. Not medical advice.

Coaches dating messages, mixed signals, conflict repair, boundaries, hard talks, and breakups with practice drills. Skills coaching, not therapy.

Runs mock interview practice with rubric feedback, builds STAR/CAR story banks, and drills system-design and coding-interview structure.

Coaches the employee and buyer side of salary, rent, bills, and big purchases with BATNA prep, verbatim scripts, and role-play practice.

Audits a tailored resume and cover letter: scores ATS match, truthfulness against the master resume, and conventions, then returns a pass/fail verdict.

Helps you stop procrastinating and beat overwhelm with GTD capture, the Eisenhower matrix, time-blocking, Pomodoro, and one concrete next action.

Public speaking coach: structures talks (SCQA, Story Spine, Sparkline), drills vocal delivery, and eases speaking anxiety with reappraisal and graded exposure.

Turns vague goals into scheduled, accountable action using the GROW model, WOOP, and if-then plans. Coaching, not therapy.

Supports everyday stress, overthinking, low mood, and burnout with evidence-based techniques offered as invitations. Not therapy; crisis-aware.

Builds good habits and breaks bad ones with cue-based design, Atomic Habits, Tiny Habits B=MAP, stages of change, and relapse prevention.

Tailors a master resume to a job description and writes the matching cover letter: ATS-optimized, country-adapted, and never fabricated.

Explores codebase, recalls memories, and narrows entry points. Outputs a refined task with <context> file references for downstream developer agents.

Turns a full spec or a rough ask into a phased, parallelizable implementation plan with disjoint task ownership, merge gates, and spec coverage.

Designs risk-based test plans with boundary, equivalence, and state-transition techniques, runs what's runnable, and writes bug reports precise enough to fix.

Builds the project's long-term memory by anchoring every entry to verified code references, then refines it with user corrections.

Turns a task into a traceable EARS spec: numbered requirements, explicit constraints, scenarios, and a verification command, with no unstated assumptions.

Generates detailed, user-facing release notes from a commit/version range by reading code, not commit messages.

Writes behavior-proving tests in any language, closes risk-ranked coverage gaps, sets up mutation tooling, and proves every assertion fails first.

Writes the project README after exploring the codebase: detects project type and audience, verifies every command by running it, and keeps sections scannable.

Reconstructs the task behind a commit, branch, or diff: the prompt as it likely arrived, plus a clarified spec with every hunk bucketed by intent.

Audits docs for staleness against code and git history, syncs drifted claims with minimum-diff edits, and writes new docs only when something real is missing.

Controls delivery without editing code: grounds architecture in source, isolates writes in Git worktrees, delegates implementation, and verifies integration.

Generates a Keep a Changelog 1.1.0 CHANGELOG.md from a commit range, verifying every entry against current code instead of commit subjects.

Implements bug fixes, features, and refactors: researches upstream fixes, surveys every affected surface, fixes root causes, and verifies with tests.

Turns a diff, commit, or range into plain-language review cards: behavioral impact, traced risk, and source links behind every finding.

Playwright browser automation: drives a headless browser to navigate pages, scrape structured data, fill forms, capture screenshots, and save artifacts to disk.

Generates or updates a thin, standard-compliant AGENTS.md: discovers the project, interviews only for non-inferable knowledge, moves depth into .agents/.

Drafts the messages people actually send: emails, complaints, appeals, announcements, and hard conversations, in your voice and ready to send.

Routes every request to the right specialist agent, remembers the user across sessions, and synthesizes each result into one answer.

Turns renewals, deadlines, subscriptions, paperwork, and appointments into one maintained ledger with lead times, next actions, and calendar reminders.

Turns meeting transcripts and notes into minutes with decisions and owned action items, then tracks follow-through across a searchable archive.

Curates the project's durable knowledge base: writes structured .box/ notes that octobrain indexes, and answers questions from them with cited sources.

Investigates questions with current web sources, cross-checks claims across independent sites, and returns structured findings with citations.

Handles everyday tasks across files, shell, web search, planning, writing, and analysis, delegating specialist work to domain agents.

Runs standing competitor monitoring, market, and topic watches: sweeps sources, verifies changes against memory, and delivers evidence-graded digests.

Triages email and Slack, drafts replies in your voice, and proposes meeting slots with timezone math. Sends only on your explicit approval.

Builds production LLM applications: RAG, agent tool schemas, context engineering, prompt caching, tracing, and MCP integration.

Designs LLM evaluation suites: golden datasets, RAGAS and G-Eval metrics, statistical significance testing, drift detection, and CI regression gates.

LLM red teaming for apps and agents: prompt injection audits, jailbreak testing with garak and PyRIT, OWASP LLM and Agentic Top 10 mapping, and defense design.

Classifies AI systems under the EU AI Act, maps obligations by tier and role, and builds an AI governance framework on NIST AI RMF and ISO 42001.

Covers Indian central law from India Code: Constitution, BNS/BNSS/BSA, CPC, Labour Codes, DPDP Act. Information, not legal advice.

Explains Canadian federal law in English or French, including Charter rights; Quebec civil law routes to a specialist. Information, not legal advice.

Covers Thai law from the Royal Gazette: PDPA, Computer Crime Act, Cybersecurity Act, Civil and Commercial Code. Information, not legal advice.

Explains Australian federal law from legislation.gov.au, read at the source; state and territory law is flagged. Information, not legal advice.

Explains French law with legislation from Legifrance/DILA, covering the civil, labor, and criminal codes, company law, and RGPD. Information, not legal advice.

Explains German federal law: statutes from gesetze-im-internet.de (BGB, HGB, StGB, BDSG); case law and DSGVO/EU via web search. Information, not legal advice.

Covers English, Welsh, Scottish, and Northern Irish law from legislation.gov.uk with EU cross-references: contracts, employment. Information, not legal advice.

Covers visa and immigration for any country: categories, documents, timelines, appeals. Official sources first. Information, not legal advice.

Covers US federal and 50-state law with case law via CourtListener: contracts, employment, criminal, IP. Information, not legal advice.

Covers Singapore law from Singapore Statutes Online, read at the source: contracts, employment, criminal, company, and property. Information, not legal advice.

Guides disputes: refunds, chargebacks, demand letters, debt-collection responses, credit report disputes, and small-claims prep. Information, not legal advice.

Explains rights, obligations, legal concepts across legal systems worldwide, reviews documents, researches law via web search. Information, not legal advice.

Manages seller-side store operations: listing copy, catalog and order management, pricing against margin, and refund playbooks. Money moves wait for approval.

Turns fuzzy ideas into evidence-backed PRDs, user stories, and RICE-prioritized roadmaps, problem first, with scope cut by an explicit never list.

Plans milestones with owners and a critical path, maintains RAID logs, sweeps stale tracker items, and writes honest RAG status reports with minimum ceremony.

Manages customer success for recurring revenue: onboards to first value, scores account health, runs churn-risk save plays, and preps QBRs and T-90 renewals.

Triages email and Slack support queues by severity, drafts replies for approval, and files tickets with error-tracking evidence and SLA flags.

Suggests what to cook with ingredients you have, builds weekly meal plans with grocery lists, scales recipes, and coaches technique with food-safety hard rules.

Builds idempotent data pipelines: ingestion, layered warehouse transforms, Postgres tuning, and data quality checks that catch drift before dashboards lie.

Exploratory data analysis assistant that turns messy data into a plain answer with honest statistics, the right chart, and correlation split from causation.

Excel formula helper: fixes broken Excel and Google Sheets formulas, lookups, and pivots by diagnosing the data first, then teaches how to avoid repeats.

Writes, debugs, and optimizes analytical SQL queries by reasoning from logical processing order, join grain, and sargability instead of guessing.

Triages car symptoms, calls DIY versus mechanic, checks repair quotes, and guides used-car buys, with hard safety gates on brakes, SRS, and EV high voltage.

Builds and runs communities: Discord server structure, onboarding, moderation policy, community engagement rituals, and health metrics beyond vanity counts.

Builds day-by-day itineraries, budgets, and transport logistics, verifying prices and opening hours against current web sources. You book; it plans.

Coaches force-free training for dogs and cats, from puppy basics to reactivity, separation anxiety, and litter-box problems. Rules out medical causes first.

Triages pet symptoms and health concerns for dogs, cats, and common pets with severity tiers, toxin checks, and home-care guidance. Not a veterinary diagnosis.

Runs rent-vs-buy math, stress-tests affordability, decodes mortgages, and covers offers and tenant basics for any country. Educational, not licensed advice.

Diagnoses home problems, makes the DIY vs pro call, and walks repairs step by step with hard safety gates on gas, mains electrical, and structural work.

Diagnoses drafts with the content-audit rubric, then applies surgical fixes that preserve the author's voice; humanizes AI text on request.

Translates sense-for-sense across locales, preserving voice and structure; also validates, native-audits, or adjusts existing translations.

Coaches fiction craft without ghostwriting: structure diagnosis, character work, honest developmental critique, and story-bible continuity tracking.

Drafts original posts, threads, quote-posts, captions, DMs, and bios across X, LinkedIn, Reddit, Instagram, TikTok, and more, in the author's voice.

Writes short-form opinionated blog posts (600-1500 words) with strong hooks, a clear point of view, and light SEO; saves to disk as Markdown.

Turns raw notes, transcripts, and brain-dumps into ready posts, threads, or blog drafts through grounded dialog; hands long-form research to content:article.

Drafts public replies to comments, reviews, and forum threads with a reply-or-skip verdict first; never posts, and discloses affiliation.

Publishes approved content to YouTube, Instagram, TikTok, and X with platform-native metadata, then reports performance against account baselines.

Turns ideas and documents into presentations: narrative structure, assertion-based slides, speaker notes, and renderable Marp markdown decks.

Read-only AI-authorship check: flags AI-typical writing patterns and hard artifacts with evidence and a calibrated band, never a fake percentage or verdict.

Orchestrates content production: sets editorial strategy and voice, briefs specialist writers, gates quality through content:audit, and approves publishing.

Researches, outlines, and drafts long-form articles (1500-3000 words) with inline citations, SEO structure, and a grounding report; saves Markdown to disk.

Read-only content audit of posts, threads, blogs, and articles by the content-audit rubric: per-axis and overall scores, prioritized findings, editor handoff.

Plans product-led growth (PLG) and hybrid SaaS growth: activation milestones, PQL scoring, funnel diagnosis, growth loops, and ICE-ranked experiment backlogs.

Designs lifecycle flows and writes newsletters and campaigns with segmentation, deliverability hygiene (SPF/DKIM/DMARC), and click-based metrics.

Packages news for media coverage: press releases, researched journalist lists, personalized pitches, founder comms, and crisis response.

Owns social media channel strategy, content pillars, cadence, and the engagement loop; delegates writing and publishing to specialists and reports outcomes.

Designs affiliate, referral, reseller, and tech-partner programs: recruitment lists, incentive economics, co-marketing plans, and cohort measurement.

Builds 30-60-90 onboarding plans, review cycles, SBI feedback scripts, performance improvement plans, and HR policy drafts. Flags legal questions for counsel.

Runs hiring from role scorecard to signed offer: job descriptions, sourcing plans, structured interview kits with rubrics, and candidate communications.

Mines Reddit, Hacker News, and review sites for user pain points, maps competitor gaps, and ranks blue-ocean product opportunities with scored evidence.

Crafts brand identities for early-stage products: names, positioning statements, messaging frameworks, brand voice guidelines, taglines, and landing page copy.

Writes ad copy, visual scenarios, and creative briefs for Google, Meta, LinkedIn, X, and TikTok: 4-6 angles per campaign, every line fitted to its slot.

Stress-tests one startup idea to a GO/NO-GO verdict with problem research, TAM/SAM/SOM sizing, competitor analysis, and a risk matrix.

Writes elevator pitches, 10-12 slide pitch decks, pricing models and tier designs, sales page copy, and partnership proposals, with objection mapping.

Builds zero-budget launch playbooks: Product Hunt plans, Show HN and Reddit launches, cold outreach templates, and social proof tactics for zero-user products.

Plans go-to-market strategy: beachhead ICP, sales motion by ACV, channel mix with test budgets, launch gates, and a metric stack with kill thresholds.

Academic writing assistant that structures research papers, decodes dense published work, drafts abstracts and reviewer responses, and checks citations exist.

Builds and hardens containers: Dockerfiles, multi-stage builds, docker-compose stacks, image scanning with Trivy or Docker Scout, and non-root defaults.

Diagnoses and fixes server issues from logs and metrics: shell scripting, Docker, CI/CD pipelines, DNS and TLS, systemd, and monitoring setup.

Responds to incidents end to end: severity triage from error tracking, mitigation-first response with kubectl and docker, and blameless postmortems.

Writes and manages Terraform infrastructure: HCL modules, remote state with locking, plan and apply workflow, drift detection, and multi-cloud providers.

Manages Kubernetes workloads with kubectl and Helm: deployments, services, ingress, RBAC, network policies, autoscaling, and production cluster operations.

Explains blood test results (CBC, metabolic, lipid, thyroid), flags out-of-range values, and surfaces patterns across markers. Information, not a diagnosis.

Explains drug mechanisms, side effects, interactions, and contraindications, and suggests questions for a prescriber. Information, not personalized advice.

Guides evidence-based sleep improvement with CBT-I-informed structure, sleep science, and diary-driven coaching. Screens red flags for sleep medicine.

Medical bill review: audits itemized bills for errors, builds dispute letters and negotiation scripts, and navigates charity care and No Surprises Act rights.

Answers health questions from current medical literature, explaining conditions and treatments in plain language. Information, not a substitute for a clinician.

Explains nutrients, dietary patterns, and supplement evidence from peer-reviewed research. Educational information, not personalized diet advice.

Triages alerts, logs, and phishing emails, prioritizes scanner findings by exploitability, and investigates incidents with evidence-cited timelines.

Reviews code and config against the OWASP Top 10:2025 and ASVS 5.0: vulnerability assessment, secure-coding fixes, threat modeling, supply-chain audits.

Maps GDPR, CCPA, and SOC 2 obligations to what a team must do: data maps, lawful bases, DPIAs, policy drafts, and questionnaire answers. Not legal advice.

Checks suspicious texts, calls, links, and offers before you pay, and guides first-hour recovery after a scam, hack, or identity theft.

Produces end-to-end short-form ads for TikTok, Reels, and Shorts: script, storyboard, generated clips, voiceover, captions, and a stitched platform-spec render.

Produces avatar talking-head UGC ads for Meta and TikTok: HeyGen renders, ElevenLabs voice, lipsync polish, b-roll cutaways, and 3-5 hook variants per campaign.

Writes video scripts for short-form and long-form: hook variants, timestamped beats, voiceover lines, on-screen text, and b-roll cues. Text artifacts only.

Plans beat-by-beat storyboards: shot list, framing, motion, transitions, music brief, and asset checklist, with optional generated reference frames per beat.

Produces 60-180s explainer videos for SaaS demos, hero videos, and YouTube pre-roll using Remotion motion graphics, voiceover, and screen-capture briefs.

Publishes finished cuts to TikTok, Instagram, and YouTube with spec validation, compliance checks, per-platform metadata adaptation, and a publish report.

Routes any video brief to the right specialist (script, storyboard, adcraft, ugc, explainer, publish), delegates phase by phase, and reconciles outputs.

Researches products against your needs and budget, checks current prices across retailers, and returns a shortlist with honest trade-offs. No affiliate bias.

Coaches SAT, ACT, GRE, and GMAT prep with retrieval practice, spaced study, official timed tests, and an error log that tags every miss by root cause.

Teaches programming through scaffolded hints instead of finished answers: PRIMM exercises, debugging as a skill, and an escape route from tutorial hell.

Coaches foreign-language self-learners toward conversation: comprehensible input, high-frequency vocab with SRS, output practice, and honest CEFR milestones.

Tutors math from arithmetic to calculus with Polya's four steps, worked examples, and Socratic questions. Guides the struggle instead of giving answers.

Tutors concepts in any subject, from sciences to writing, using retrieval practice, Socratic questions, and spaced review. Guides reasoning; never ghostwrites.

Coaches budgeting, debt payoff, and emergency funds using frameworks like 50/30/20 and snowball vs avalanche. Education, not personalized financial advice.

Explains how personal income tax works in any country: brackets, freelancer obligations, capital gains, and moving abroad. Rates come from official sources.

Interprets financial statements, computes ratios, explains valuation methods like DCF and comps, and spots red flags. Educational, not investment advice.

Guides policy choice, decodes coverage and exclusions, structures claims, and drafts denial appeals with ombudsman escalation paths. Educational, not a broker.

Trades Binance USDT-M futures: reads market structure, sizes positions from account risk, and manages entries, stops, and exits. Orders need confirmation.

Keeps books for freelancers and small businesses: invoicing and AR follow-up, processor reconciliation, monthly close, and a 13-week cash-flow view.

Decodes any country's pension system into three pillars, runs on-track math, and covers late-start playbooks and scam radar. Educational, not regulated advice.

Drives mobile and desktop UIs through reproducible flows, capturing checkpoint screenshots and a run.json step log as evidence.

Guides adult children through an aging parent's care: ADL/IADL assessment, the care ladder, dementia caregiving, conversation scripts, and funding patterns.

Supports parents from newborn to teens on sleep, tantrums, discipline, screens, and teen conflict with evidence-based, culture-aware guidance. Not diagnostic.

US benefits and job-loss navigator: screens eligibility via official tools, builds document checklists, tracks deadlines, and preps renewals and appeals.

Guides the weeks after a death: the immediate checklist, notifications, estate orientation, scam radar, and grief-aware pacing. Educational, not legal advice.

WCAG 2.2 AA, EAA, and ADA Title II accessibility audits: rates barriers by user impact; delivers audit reports, statements, and remediation roadmaps.

UX audits of pages and flows against Nielsen's heuristics and WCAG 2.2 AA, with findings rated by severity and failing microcopy rewritten in place.

Turns interview transcripts and feedback into coded themes where every claim traces to verbatim quotes, plus personas and a queryable insight repository.

Designs revenue systems: enrichment waterfalls, signal-driven outbound plays, lead scoring and routing, and deliverability setup. Specs systems, never sends.

Executes the sales motion: lead research, CRM hygiene, outreach drafts for email and LinkedIn, and meeting setup. Every send waits for approval.

Coaches B2B cold outreach: rewrites email, LinkedIn, and call scripts using REPLY and 4-T, designs multi-touch cadences, and flags deliverability problems.

Turns RFPs into compliant proposals: bid/no-bid scoring, compliance matrices, win themes with proof, and evaluator-focused drafts with color-team review.

Coaches B2B closing and negotiation, turning stalls and indecision into signed deals with JOLT, mutual action plans, and verbatim phrasing. Protects margin.

Coaches B2B discovery and qualification: uncovers quantified pain with SPIN, scores deals against MEDDPICC, and frames demos around the buyer's problem.

Octomind built-in core server: recall of folded context blocks. Planning is supervisor-owned, not a tool.

UK legislation search, point-in-time versions, and amendment tracking via the legislation.gov.uk MCP server.

High-trust Octomind runtime controls: MCP server management, dynamic agents, and skill/capability activation.

Text-to-speech and audio transcription via the Whisper MCP server (OpenAI tts-1).

Geocoding, directions, places, and elevation via the Google Maps API (mcp-google-map server).

URL fetching with markdown conversion for LLM consumption, via the fetch MCP reference server.

Google Calendar event and schedule management via the @cocal/google-calendar-mcp server.

Official Svelte and SvelteKit documentation via the @sveltejs/mcp server.

Browser automation via Octoweb local HTTP MCP server.

Slack messaging - channels, messages, threads, and users - via the official Slack MCP server.

Python package and project management via uv and uvx.

Stripe customer, charge, subscription, and invoice management via the official @stripe/mcp server.

Kubernetes cluster management via kubectl and helm CLI tools.

Instagram Reels, feed video, and carousel publishing with media insights via the Meta Instagram Graph API (@mcpware/instagram-mcp).

Lip-sync video generation and postprocessing via the Sync.so MCP server (sync-1.9 / sync-2 models).

YouTube video metadata, SEO, and channel analytics after upload, via the YouTube Studio MCP server (OAuth2).

Shopify store management via the Shopify Dev MCP server: products, orders, customers, inventory, and storefront.

US case law search and opinions via the CourtListener (free.law) MCP server.

Rust toolchain with cargo, rustc, clippy, and rustfmt.

Stock quotes, financial reports, and market data via the yfinance MCP server (Yahoo Finance).

French legislation from Legifrance/DILA via the Ansvar French law MCP server.

Code relationship and dependency-graph traversal via octocode: graphrag.

Error tracking - issues, events, and releases - via the Sentry MCP server.

File editing via octofs: text_editor, batch_edit, and extract_lines; pairs with filesystem-read.

Issue tracking and project management via the Linear MCP server.

Semantic memory recall via the octobrain MCP server (remember tool, with 1-hop graph neighbors).

Read-only file and directory access via octofs: view, workdir.

Docker container management via docker and docker-compose CLI.

PostgreSQL database querying and inspection via the official MCP reference server.

GitHub repositories, issues, pull requests, and branches via the official GitHub MCP server.

Version-specific library and framework documentation lookup via Context7.

TikTok video and photo uploads (direct post or inbox), scheduling, and status tracking via the @upload-post/mcp gateway (TikTok Content Posting API).

Fast audio transcription (speech-to-text) and conversion via the Whisper MCP server.

Symbol and AST code search via octocode: structural_search, view_signatures.

Text-to-image generation (Flux Schnell) and SVG (Recraft V3) via Replicate.

Drug information, clinical data, and guidelines from PubMed, FDA, WHO, and RxNorm via the medical MCP server.

German federal statutes fetched live from gesetze-im-internet.de via the german-law MCP server.

Semantic search over the project's persistent knowledge base: local .box docs, subscribed org boxes, and indexed files, scoped to the active project.

WhatsApp Web messaging, contacts, and media sharing via the @iflow-mcp/mcp-whatsapp-web MCP server.

Hedra Character-3 image-to-talking-character video, text-to-speech, and image generation.

Memory creation and deletion via the octobrain MCP server (memorize and forget tools).

Instagram messaging - posts, comments, DMs, stories, and reels - via the @mcpware/instagram-mcp server.

Deep web scraping, crawling, and site mapping via Firecrawl.

Telegram bot messaging - messages, channels, and groups - via the @iflow-mcp/telegram-mcp-server.

iOS and Android device and simulator automation via the Mobile MCP server.

Cross-domain agent delegation (tap) and session-flow control (schedule) via the built-in orchestration server.

Text-to-video and image-to-video via Runway Gen-4, with video upscaling and editing.

Binance USDT-M futures trading: candles, order book, funding, positions, orders, leverage, and margin.

Bun all-in-one JS/TS runtime (bun, bunx) - drop-in for node, npm, npx, yarn, pnpm, and vite.

Drive the local computer UI — mouse, keyboard, screenshots — via RobotJS (macOS / Linux / Windows).

SMS and voice messaging via the Twilio MCP server (@twilio-alpha/mcp, twilio_api_v2010 service).

Natural-language code search via octocode: semantic_search.

Reading highlights, book annotations, and saved articles via the Readwise MCP server.

Swift toolchain for iOS, macOS, and server development: swiftc compiler, REPL, and Swift Package Manager.

Drive the Octoweb browser (tabs, navigation, snapshots, clicks,

Web search via the Brave Search API (brave_web_search tool).

Local SQLite database querying and management via the official MCP reference server.

HubSpot CRM customer management - contacts, deals, and campaigns - via the official HubSpot MCP server.

Task delegation to configured sub-agents via the built-in agent server.

Video composition via ffmpeg, installed locally and invoked through the shell tool.

Stock images, vectors, and videos - search, download, and AI image generation - via the Freepik API.

Discord messaging - channels, messages, moderation, and voice - via the mcp-discord server.

Cloudflare Workers, KV, R2, D1, and DNS management via the Cloudflare MCP server.

ElevenLabs music composition, video-to-music, and sound effects via the ElevenLabs MCP server (elevenlabs-mcp).

Shell command execution via the octofs MCP server (octofs:shell tool).

X post creation, deletion, and analytics via the official hosted X MCP (@xdevplatform/xurl bridge), with media upload and recent search.

Email messaging over standard SMTP and IMAP protocols via the @codefuturist/email-mcp server (works with any mail server).

Canadian federal legislation with bilingual support via the Ansvar Canadian law MCP server.

Catalog of proven first-1.5-second hook patterns for short-form video (TikTok, Reels, Shorts): an 11-pattern hook catalog (pattern-interrupt, contrarian, problem-aware, curiosity-gap, stat punch, identity bait, and more) plus visual hook devices that stack on top of the script, the rules for each, and the banned opens that tank retention. Use whenever drafting an ad or social-video script; the hook is what decides whether the rest of the clip ever gets watched.

Operational funnel for discovering high-intent, low-competition long-tail keywords (4+ word phrases with weak SERP defenders), the keyword class that actually ranks for new and mid-authority sites without massive backlink spend. Encodes the discovery method (seed to modifier expansion to Google autocomplete + 'People Also Ask' + Reddit/forum mining + competitor SERP-gap), the qualification rubric that goes beyond KD scores (top-10 SERP authority profile, content-fit gap, intent match, traffic-potential vs ranking-realism, query freshness), the 2026 GEO overlay (definition + question + comparison queries that earn AI-search citations), and the kill-rules (head-term traps, fake-volume keywords, branded-SERPs, paid-ads-dominated SERPs). Use BEFORE writing any content brief, calendar, or pillar-cluster plan. Stays in the SEO lane: produces a keyword brief; content writing, on-page wrapping, and link-building are downstream concerns owned elsewhere.

Operational playbook for designing production LLM agents in 2026. Covers agent loop patterns (ReAct, Reflexion, Plan-and-Execute, native tool use), tool schema engineering (descriptions, strict mode, input examples, tool search, anti-patterns), the Agent Skills open standard, Model Context Protocol (spec 2026-07-28) integration and security, multi-agent orchestration (LangGraph, CrewAI, AutoGen, vendor agent SDKs) with cost trade-offs, error handling, loop prevention and unattended-run termination, observability via OpenTelemetry GenAI semantic conventions, OWASP LLM and Agentic Top 10 mapping, and context engineering (just-in-time retrieval, compaction, memory, sub-agent isolation, caching). Use when designing a new agent, picking an orchestration framework, debugging a misbehaving agent, or hardening tool exposure. Output: agent specs with named patterns, cost projections, observability hooks, and OWASP-mapped risk surfaces.

Write and revise Meta Threads posts, replies, and launch copy in the author's documented voice, covering format choice, honest proof, conversation, and conversion paths. Use for posting on Threads, a Threads campaign, or a threads.com brief; excludes multi-post threads on other platforms.

Per-language calibration anchors for detecting AI-slop and translationese in Portuguese (pt) target text; auto-loads alongside content-locale-humanize when the target language is Portuguese.

Country-by-country conventions for job-application documents in 2026: resume vs CV terminology, length, whether to include a photo, data-processing consent clauses, work-authorization lines, date and locale formatting, and personal-data norms. Adapt an application to the target market so it reads correctly and stays legally safe wherever it's sent.

Specifies what to produce for a Meta ad: Facebook Feed, Instagram Feed, Stories, Reels, and the CTA enum, with the exact slots (primary text, headline, description, CTA button), character limits, asset ratios (1:1, 4:5, 9:16), policy filters, and the truncation behavior that decides whether anyone reads past the fold. Use whenever the task is producing Meta ad copy or asset briefs; strictly platform-specific, so it does not cover audience setup, pixel events, or campaign objectives.

Modern C# and .NET best practices for AI coding agents: nullable contracts, async ownership, data access, and explicit failure handling. Auto-activates in .NET projects.

The small set of evidence-based study techniques that actually work: retrieval practice, successive relearning, spacing, interleaving, self-explanation, and worked examples, plus how to use AI tutors without outsourcing the thinking, and the popular habits (rereading, highlighting, cramming, learning styles) that only feel productive. Activate when helping someone study, memorize, prepare for a test, or learn a skill, or when they ask how to learn something faster.

Harvest current Meta Threads conversations into a source-backed brief with post URLs, visible metrics, audience questions, and reusable structures. Activate for Threads trend scans or campaign research; distinguish observed conversation from inferred reach and buying intent.

Reference for octomind's out-of-band control plane: the self-report token, deterministic loop/no-progress/recovery detectors, the verify-gate on every turn stop, the supervisor-owned plan manager, condense, steer, recite, the optional tool authorizer, calibrated evaluation gates and hindsight, and cross-session lessons, orientation, and experience memory with optional evolution, together with the strict [supervisor] config sections. Activate when a user asks why a turn was re-run or gated, how the detectors work, why a tool call was blocked, how lessons are stored and recalled, why tool output was condensed, or how to tune any [supervisor.*] section.

Svelte 5 and SvelteKit best practices for AI coding agents: runes, architecture, server/client boundaries, and progressive enhancement. Auto-activates in Svelte projects.

Operational skill for landing genuine editorial guest posts that earn high-quality contextual links without tripping Google's link spam and scaled content abuse policies. Encodes the 7-stage workflow (author tier-mapping, discovery operators, 11-point site qualification, content-gap topic ideation, 80-150-word pitch architecture, post writing standards by tier, internal-link strategy, anchor distribution, follow-up). Bakes in Google's hard kill-rules (paid placements, optimized anchors in guest posts, duplicate content, AI-content farms), a practitioner anchor-distribution heuristic (branded majority, exact-match minority), and link-placement rules (1-2 contextual body links + 1 bio). Use BEFORE writing any pitch. Stays in the SEO lane: produces strategy, target list, topics, pitch brief; does not write the email or post itself.

Harvest X conversations into a sourced topic brief. Activate when asked to scan Twitter or X trends, buyer pains, launch reactions, or post formats; validate live surfaces, compare relevant accounts with explicit metric limits, and distinguish observed patterns from ranking hypotheses.

Generate, improve, analyze, and debug LLM prompts. Activate only when the user explicitly asks to write, rewrite, optimize, diagnose, or design a prompt/system prompt, or names a prompt-engineering technique.

Per-language calibration anchors for detecting AI-slop and translationese in Chinese (zh) target text; auto-loads alongside content-locale-humanize when the target language is Chinese.

Per-language calibration anchors for detecting AI-slop and translationese in Korean (ko) target text; auto-loads alongside content-locale-humanize when the target language is Korean.

Per-language calibration anchors for detecting AI-slop and translationese in Hindi (hi) target text; auto-loads alongside content-locale-humanize when the target language is Hindi.

Diagnose and maintain Fedora systems with release-aware DNF5, Atomic deployment, SELinux, networking, and rootless container guidance.

The dimensions of a deep code review: correctness, security, concurrency, performance, design/maintainability, and tests; what each lens hunts for; and the adversarial verification discipline that keeps a multi-reviewer pass from multiplying false positives. Consensus is not correctness; every finding must be refuted before it is trusted.

Diagnose and maintain Red Hat Enterprise Linux hosts and images with release-aware repositories, networking, SELinux, and container operations.

Pythonic best practices for AI coding agents: architecture, type-driven design, and library choices that scale beyond scripts. Auto-activates in Python projects.

Operational funnel for finding and qualifying SAFE free places to place backlinks. Encodes the prospect categories (reclamation, resource pages, broken links, journalist requests, communities, aggregators, niche directories), the per-target qualification rubric (topical fit, real traffic, outbound profile, page context, anchor naturalness, reachability), Google's link-spam guardrails (penalty triggers, rel attribution rules), and the 2026 AI-search citation overlay (Reddit/community as Perplexity/ChatGPT citation sources). Use AFTER niche/audience are clear and BEFORE any outreach. Stays in the SEO lane: produces a prospect brief; outreach copy and on-page wrappers are downstream concerns owned elsewhere.

Operational playbook for diagnosing the structural and technical layer of a site or page: crawlability (robots.txt), indexability (meta robots, canonicals), sitemap.xml validity, redirect chains, JSON-LD schema validation, Core Web Vitals interpretation, hreflang, JS-render parity, URL architecture, and internal-link topology. Encodes the fetch protocol (parallel webfetch + shell curl), the per-surface check rubric, the severity matrix (Critical / Moderate / Minor), and the unverified-check escalation rule. Use when auditing the technical layer of a URL, sitemap, or site. Output: prioritized structural findings with evidence and fix direction.

MQM-informed, structural-pattern diagnostic for making translated/localized content read as if a native speaker wrote it from scratch, in any target language. Detects translationese and AI-slop by pattern (interference, frequency outliers, rhythm, templating), not by a per-language word list.

Draft and revise Reddit posts, comments, AMAs, and product launches in the author's documented voice. Use when preparing content for a subreddit: check community rules, choose a native format, ground claims in supplied proof, disclose affiliations and AI use, and plan useful replies and outcome measurement.

Modern C++ best practices for AI coding agents: architecture, RAII, value semantics, and reproducible dependency and library selection guidance. Auto-activates in C++ projects.

Operate the local computer's native UI (macOS, Linux, or Windows) via the desktop-automation MCP server. Covers OS accessibility and screen-recording permissions, pixel-anchor targeting (no semantic UI tree), full-screen capture limits under the MCP response cap, HiDPI coordinate mapping, window focus and multi-monitor handling, keyboard and mouse sequencing, and the safety boundary around destructive desktop actions and screen-lock or biometric bypass. Activate when the user wants to drive a native desktop app or the OS shell at the GUI layer. Pairs with the device:operator agent, which carries the shared snapshot, verify, and record discipline; this skill carries the desktop-specific tooling and rules and pulls the device-computer capability automatically.

Specifies what to produce for a Reddit ad: Promoted Posts (image, video, link, text), Free-form Ads (the rich-content flagship, launched 2024), Carousel Ads, Conversation Placement Ads, Promoted User Posts (boost an existing organic), Product Catalog Ads, and AMA Ads, with the exact slots, char limits, asset ratios, CTA enum, and the Reddit-specific creative rules (native voice, per-subreddit variants, real-artifact imagery, comment engagement plan) that decide whether the ad performs or gets downvoted into oblivion. Use whenever the task is producing Reddit ad copy or assets; strictly platform-specific, so it does not cover subreddit targeting strategy or Reddit Pixel setup.

Idiomatic Rust best practices for AI coding agents: architecture, ownership patterns, and ecosystem choices that survive long-term maintenance. Auto-activates in Rust projects.

Validation funnel for deciding which free tool to build for backlinks, embeds, and AI-search citation, encoding the data-moat-first methodology: niche-fit gate, data inventory, 5-candidate scoring across niche/moat/SEO/feasibility/monetization, alignment checklist, and a strategy brief. Use before any tool-build work to avoid commodity tools that earn nothing; skip if the user has already validated the concept and just wants to build, since the deliverable is the brief and the build itself is downstream.

Guide for octomind workflow files: pipeline design (pattern selection, step decomposition, sentinel handoff contracts, loop convergence, context economics) plus the exact TOML syntax for the external orchestrator that chains octomind run subprocesses. Covers sequential, parallel (static, count, dynamic fan-out), loop, conditional, and graph-routed steps, variable substitution, session modes, pre-flight validation, tap workflows, and the CLI. Activate when designing, authoring, validating, or debugging a workflow .toml file.

Compact playbook of ad copy frameworks (AIDA, PAS, problem-agitation-solution, before-after-bridge, 4Us, FAB) and when to apply each, including a decision guide that matches framework to audience awareness level (Schwartz). Use when structuring the body of an ad: script, body copy, landing-page sections, or any persuasive piece.

Read-only audit harness for content quality: detects the content type (X post / X thread / LinkedIn / Threads / Bluesky / Mastodon / Hacker News / Reddit post, or blog / article), routes to the matching social-* skill plus the cross-cutting voice/humanize/grounding/geo checks, scores per-axis 0-10 + overall 0-100, and emits a canonical findings report with severity tiers, evidence, and suggested rewrites. For X surfaces also emits a Phoenix-style verdict (Pass / Borderline / Likely-Suppressed / Kill-Switch-Risk); single source of truth for the diagnostic shape, used by content:audit (read-only) and content:editor (diagnose-then-edit).

Operational playbook for designing and tuning production RAG (Retrieval-Augmented Generation) systems in 2026. Covers chunking strategies (fixed/recursive, semantic, late chunking, Anthropic Contextual Retrieval), retrieval (BM25, dense, hybrid with reciprocal rank fusion, ColBERT late-interaction), query rewriting (HyDE, multi-query, decomposition), reranking (Cohere Rerank 4, cross-encoders, BGE-Reranker), evaluation (RAGAS metrics, TruLens RAG Triad, DeepEval), the seven failure modes (Barnett et al.), agentic RAG and agentic search, multi-modal RAG (ColPali), and the long-context-vs-RAG trade-off including context rot. Use when designing a new RAG pipeline, diagnosing why an existing one underperforms, or evaluating retrieval quality. Output: architecture decisions with cited numbers, not vibes.

Per-language calibration anchors for detecting AI-slop and translationese in Italian (it) target text; auto-loads alongside content-locale-humanize when the target language is Italian.

Per-language calibration anchors for detecting AI-slop and translationese in Arabic (ar) target text; auto-loads alongside content-locale-humanize when the target language is Arabic.

Specifies what to produce for a TikTok ad: In-Feed Ads, Spark Ads, TopView, Branded Effect, and Branded Mission, with the exact slots (caption, display name, CTA), character limits, the 9:16 vertical asset spec, sound-on requirements, the safe-zone map that decides what is visible on phones, and the native-feeling creative rules that decide whether the algorithm pushes or buries. Use whenever the task is producing TikTok ad copy or video briefs; strictly platform-specific, so it does not cover audience setup or pixel events.

Windows client/server operations: release channels, PowerShell context, services, updates, permissions, WSL boundaries, and evidence-based recovery.

Harvest Bluesky conversations, formats, and launch evidence into a sourced brief. Activate when scanning Bluesky trends or researching post angles, custom feeds, and relevant community responses; record observed surfaces and account baselines without inventing amplification paths.

Draft and revise TikTok scripts, captions, and publishing packages in an identifiable author's voice. Activate for TikTok posts, product demos, launches, customer proof, and creator collaborations; choose formats by audience intent and separate verified platform guidance from editorial judgment.

Diagnose and maintain Debian systems: release support, APT sources, package failures, service persistence, networking, and supported upgrades. Use for Debian hosts and Debian-based images.

Harvest YouTube topic demand, channel-relative outliers, buyer questions, and reusable format ideas. Activate for YouTube trend scans, competitor or launch monitoring, and evidence-backed content briefs; separate observed public data from private analytics and turn findings into original angles with explicit limitations.

Operational playbook for defending LLM applications and agents against prompt injection (direct + indirect) and red-teaming them in 2026. Covers the OWASP Top 10 for LLM Applications (2026 edition, with 2025 ID mapping), the OWASP Top 10 for Agentic Applications (ASI01-ASI10), foundational research (Greshake et al., Many-Shot Jailbreaking, Crescendo), documented exploits (EchoLeak CVE-2025-32711, GitHub Copilot CVE-2025-53773, ForcedLeak, Slack AI), architectural defenses (lethal trifecta / Rule of Two, CaMeL, injection-resistant design patterns), defense layers (tagging and spotlighting, structured outputs, guard models such as Llama Guard 4 / Prompt Guard 2 / gpt-oss-safeguard, NeMo Guardrails, least agency, MCP hardening), and red-team tooling (garak, PyRIT, Promptfoo). Use when auditing an LLM app for injection, designing defense-in-depth, mapping OWASP risks, or building a red-team suite. Output: red-team findings + defense recommendations with reproducible payloads.

Operational playbook for cutting LLM application cost in 2026. Covers the current price book (Anthropic Claude Fable/Opus/Sonnet/Haiku, OpenAI GPT-6 Astra/Sol/Luna, Google Gemini Pro/Flash/Flash-Lite), prompt caching mechanics (Anthropic write/read multipliers and TTLs, OpenAI and Gemini automatic caching), Batch and Flex tiers (50% off), effort and reasoning-depth controls, model routing vs effort tuning, output-length discipline, the RAG cost stack (embeddings, rerank, vector DBs), and FinOps attribution (provider usage APIs, Helicone, Langfuse, Phoenix, Vantage). Use when projecting LLM cost, hunting waste in an existing pipeline, picking a model or effort level, or setting up per-feature attribution. Output: cost-per-task projections with cited, dated prices and quantified optimization levers.

Modern idiomatic Java best practices for AI coding agents: records, sealed types, pattern matching, virtual threads, and explicit failure contracts. Auto-activates in Java projects.

The universal coaching method any coach agent runs: GROW, ICF active listening, open non-leading questions, goal-setting with WOOP and if-then plans, and accountability design. Activate when coaching someone toward a goal, running a 1:1, or designing follow-through, and when you catch yourself dispensing advice instead of drawing the answer out.

Idiomatic Ruby and Rails best practices for AI coding agents: objects, services, ActiveRecord persistence, transactions, and Active Job boundaries. Auto-activates in Ruby projects.

Write a cover letter that a hiring manager actually reads: grounded in the candidate's real experience, tailored to the specific posting and company, structured for a hiring-manager skim, and free of the generic AI-template phrasing that gets letters discarded. Truthful, specific, and human.

Review and revise stiff or formulaic writing with an evidence-based eight-dimension editorial diagnostic and seven-pass edit. Preserve facts, author voice, and platform authorship rules; no detector-evasion promises.

Write and revise Instagram captions, Reels, carousels, Stories, and launch or proof posts in the supplied author's voice. Activate when drafting for Instagram; choose formats, build a usable publishing package, and connect attention to a relevant next action with evidence-backed platform guidance.

Harvest Reddit conversations and community-specific opportunities with cited posts, current rule checks, and explicit coverage limits. Activate when researching Reddit trends, scanning subreddits, or preparing a Reddit evidence brief; return observed patterns and qualified angles for downstream drafting.

Git commit message conventions, branch naming, and workflow best practices for clean version history. Activate when committing, branching, rebasing, or reviewing git history.

Canonical reference for platform-by-platform video output specs: aspect ratio, length cap, codec, bitrate, frame rate, color, caption style and safe-zone bands. Per-platform spec table covers TikTok, Reels, Shorts, IG/FB feed, Stories, Snapchat, YouTube long-form, X, LinkedIn, and Pinterest, plus safe-zone reserves, caption rules, and copy-paste ffmpeg presets. Use when picking output settings, validating an export, or spec-checking before publishing.

Write and revise DEV articles, discussions, comments, and launch or proof material in the author's documented voice. Activate for dev.to drafts, publishing packages, series, challenge entries, or company cross-posts; apply current disclosure and promotion rules before preparing copy for publication.

Operational playbook for recurring social-account hygiene executed via a headless browser: profile/bio sync across networks, cross-posting a pre-written piece to N platforms, scheduled-post audits, follower/connection cleanup, notification triage, archive/delete of old posts — within each platform's automation terms. Activate only when the user wants the browser to drive social account ops as maintenance work. Authoring new post copy is out of scope; text must be supplied as input.

Guide for writing Octomind agent manifests: TOML format, required and forbidden fields, capability tiers, system prompt structure and tone for current models, sampling and model settings, placeholders, lint rules, and the pre-write checklist. Activate when creating or editing agents/<domain>/<spec>.toml files.

Plan cross-platform social launch, proof, and objection campaigns around buyer needs, sourced evidence, channel roles, reply ownership, and business outcomes. Activate for launch planning, go-to-market strategy, GTM, or social campaigns.

How modern applicant tracking systems (ATS) rank resumes in 2026 and how to tailor a resume per job posting so it clears them: keyword strategy grounded in the candidate's real experience, job-title placement, coverage targets, parser-safe formatting, knockout questions, and the anti-stuffing and hidden-text rules that screeners and recruiters now catch. Truthful tailoring only.

Harvest TikTok trend evidence into a source-linked brief with audience fit, observed momentum, original angles, and commercial-use constraints. Activate for TikTok trend scans, Creative Center research, and browser-based campaign discovery; separate observed posts from ad examples and generated suggestions.

Idiomatic Kotlin best practices for AI coding agents: null-safety, coroutine ownership, Flow delivery, and Android/Multiplatform boundaries. Auto-activates in Kotlin projects.

Write and revise X posts, replies, threads, profiles, and launch proof in the author's documented voice. Activate for X or Twitter publishing copy, product announcements, demos, customer outcomes, and post-publication reply plans, choosing formats and conversion paths using dated evidence without inventing experience or promising reach.

Specifies what to produce for a Google Ads creative: Search RSA, Performance Max, and Display, with the exact slots (headlines, descriptions, paths, long headline, long description), character limits, pinning rules, asset ratios, and policy constraints so copy fits before it is pasted into Google Ads Editor. Use whenever the task is producing Google Ads ad copy or assets; strictly platform-specific, so it does not cover bidding, keywords selection, audience setup, or attribution.

Harvest Mastodon posts and topic conversations into a sourced, server-scoped brief. Activate for Mastodon or Fediverse trend scans and harvesting requests; verify accessible surfaces, preserve sampling limits, and distinguish observed participation from inferred spread or writing recommendations.

Complete guide for writing dep scripts in the octomind-tap registry: required header comments, platform.sh boilerplate, type: mcp vs type: dep, npx/uvx MCP pattern, platform variables and functions, companion .md format, and validation. Activate when creating or editing deps/<org>/<tool>.sh files.

Verify factual claims, preserve source meaning during edits, and separate publishable copy from evidence notes. Use for product, technical, customer-proof, and current-event content.

Per-language calibration anchors for detecting AI-slop and translationese in French (fr) target text; auto-loads alongside content-locale-humanize when the target language is French.

Write and revise LinkedIn posts, launch copy, comments, and supporting profile or long-form copy in the author's documented voice. Activate for LinkedIn drafting, with evidence-led format selection, customer-proof handling, and a complete publishing package.

Harvest Hacker News discussions and inspectable artifacts into a dated, source-linked brief. Activate when researching HN trends, prior submissions, launch examples, or audience objections; separate observed attention from ranking hypotheses and product demand.

Guide for creating AgentSkills-compliant SKILL.md packs in an Octomind tap: frontmatter fields, auto-activation rules and domains, bundled resources, validate scripts, body structure, tone, and the lint gates. Activate when creating or editing skills/<name>/SKILL.md files.

Production playbook for building modern lightweight landing pages that sell. Covers automatic dark/light theming via CSS custom properties, system-font Apple-style typography, opinionated section anatomy (hero, features, comparison, pricing, FAQ, footer CTA), conversion-focused copy hierarchy, SEO/OG/JSON-LD baseline, and the structural patterns that move visitors to purchase. Activate whenever drafting, designing, or implementing a marketing or product landing page.

Operate Alpine hosts and images with musl ABI awareness, apk repository discipline, OpenRC services, and persistent configuration.

Harvest LinkedIn posts and conversations into a sourced brief for a specific audience. Activate for LinkedIn trend scans or research briefs; capture current surfaces, buyer questions, format patterns, and evidence limitations without inventing ranking weights.

Prepare source-grounded Hacker News submission briefs, human-authorship handoffs, and launch response plans. Activate for Show HN, Ask HN, Tell HN, Launch HN, linked articles, or HN comments; choose the right format, preserve author voice, and connect useful discussion to product evaluation.

React best practices for AI coding agents: architecture, hooks, state boundaries, server components, and library choices that scale. Auto-activates in React projects.

Professional voice, clear prose, and context-aware editing for posts, replies, articles, and scripts. Preserve author facts and register without formulaic rhetoric or fabricated experience.

Specifies what to produce for a Microsoft Advertising (Bing) creative: Responsive Search Ads, Multimedia Ads (with 90-char long headlines), Audience Ads (MSN, Outlook.com, Edge), Performance Max, Shopping, and Lodging/Travel verticals, with exact slots, char limits, asset ratios, CTA enum, the AI Max and Copilot creative controls, a LinkedIn Profile Targeting note, and the trademark/political/regional policy differences from Google. Use whenever the task is producing Microsoft Ads / Bing copy or assets; strictly platform-specific, so it does not cover keyword strategy, bidding, audience setup, or UET pixel.

Generative Engine Optimization (GEO) and Answer Engine Optimization (AEO) patterns for getting content cited by AI search engines. Answer-first structure, schema markup, extractable passages, and crawler setup.

How to write a behavioral spec an autonomous agent can implement with zero unstated assumptions: EARS requirement patterns, the requirements/constraints/verification triple, proof forms (scenario, property, check), wording discipline, spec layering and split rules (repo/feature/task, parent-child), divergence-based clarification for interactive and autonomous runs, and the scratchpad rewrite discipline. Activate when writing, revising, or reviewing a feature spec or acceptance criteria.

Guide for creating Octomind capabilities: the config.toml triggers file, provider files and the default.toml symlink, MCP server wiring with env and headers, version pins, setup-symlinks.sh registration, and the validation checklist. Activate when creating or editing capabilities/<name>/ files.

Harvest Instagram post, carousel, Reel, and audio patterns into a sourced trend brief with observed metrics, original adaptation angles, and uncertainty. Activate in trend sessions mentioning Instagram or Reels, or on explicit Instagram trend, scan, analysis, or harvest requests.

Identify and operate Linux hosts, VMs, and containers: distro and ABI boundaries, services, resource limits, networking, and evidence-based recovery.

Operational playbook for executing backlink submissions via a headless browser: driving directory, listing, and business-profile forms with Playwright, mapping a single site profile to per-target fields, handling signup + email verification + captchas, human-paced throttling within Google's link-spam policies, and producing an auditable submission log. Activate only when the user has a qualified prospect list and wants the browser to submit. Prospect discovery, qualification, and anchor/description copy are upstream concerns; this skill is execution-only.

Operate iOS and Android devices (real hardware, simulators, and emulators) via the mobile-mcp server. Covers device discovery and selection, app install, launch, and terminate, accessibility-tree targeting, taps, swipes, typing, and device-button presses, orientation handling, and the safety boundary around app-store, jailbreak, biometric, and in-app-purchase protections. Activate when the user wants to drive a phone or tablet (iPhone, Android, iOS Simulator, Android Emulator) through a UI flow. Pairs with the device:operator agent, which carries the shared snapshot, verify, and record discipline; this skill carries the mobile-specific tooling and rules and pulls the device-mobile capability automatically.

Modern Swift best practices for AI coding agents: value semantics, API design, isolation, task ownership, and package compatibility. Auto-activates in Swift projects.

Elixir best practices for AI coding agents: pattern matching, supervision, process-lifecycle, and concurrency design, and Phoenix conventions. Auto-activates in Elixir projects.

Explains how octomind assembles a session at runtime: tap resolution of domain:spec tags, manifest placeholders, built-in tool servers, capability activation and the deterministic embedding auto-activator, domain and env gating, LRU eviction of the tool surface, skill discovery and activation, Agent Plugins, local .agents/tools scripts, and sandbox mode. Activate when a user asks why a tool or skill did or did not appear, how a tag resolves, how capability auto-activation decides, why a capability was evicted or refused, or where skills and taps live on disk.

Field-by-field reference for octomind's config.toml: the [model] profile and its per-boundary overrides, limits and spending caps, [capabilities] and [taps] overrides, [[roles]] with [roles.model] and [roles.mcp], [mcp] servers (builtin, http, stdio), [[hooks]], [[layers]], [[commands]], [[agents]], [[prompts]], [skills], [timing], [compression], [registry], multi-file merge, project guardrails, and template variables. Activate when a user asks how to configure, tune, validate, or upgrade octomind, add a role or MCP server, set limits or spending caps, or debug a config parse error.

Specifies what to produce for a YouTube ad: Skippable In-Stream (TrueView), Non-Skippable, Bumper, In-Feed (Discovery), Masthead, and YouTube Shorts, with the exact slots (headline, description, companion banner, CTA), duration limits, asset ratios (16:9, 9:16, 1:1), the first-5-seconds hook rule that decides whether anyone watches past skip, and the policy constraints. Use whenever the task is producing YouTube ad copy or video briefs; strictly platform-specific, so it does not cover audience targeting, conversion linkage, or Google Ads campaign setup.

Type-first TypeScript best practices for AI coding agents: architecture, module boundaries, and pragmatic library choices. Auto-activates in TypeScript projects.

PHP 8 best practices for AI coding agents: write and review concise, typed PHP 8.x code with version-aware features through PHP 8.5, explicit validation, and practical quality checks. Auto-activates in PHP projects.

Code review checklist and guidelines for giving and receiving constructive feedback on pull requests. Activate when reviewing a PR, preparing code for review, or establishing review standards.

Turn raw data into a decision-ready report a human actually reads: structured by the Minto Pyramid (answer first), every number traced to the data, the right chart for each question, and delivered as a self-contained HTML document that opens in any browser and prints to clean PDF with no dependencies. The output is format-flexible, grounded, and never fabricated.

Operational playbook for training, evaluating, and publishing muvon/octomind-embed, the granite-embedding-30m WiSE-FT soup that powers octomind's capability auto-activation and skill semantic() rules. Covers when to retrain (any change to capabilities/*/config.toml triggers or skills/*/SKILL.md semantic phrases), the bin/train pipeline (LLM augmentation, dataset build, hard-negative re-mining, CachedMNRL + Matryoshka fine-tune, soup sweep, holdout and eval_real evaluation, int8 ONNX export and calibration), resuming an interrupted run, calibrating the runtime threshold and margin, publishing behind the regression gate, and recording the baseline. Use when triggers or semantic rules change, when retraining or republishing the tap model, or when auto-activation picks the wrong capability.

Specifies what to produce for an X (Twitter) ad: Promoted Posts, Website Cards, App Cards, and Video Ads, with the exact slots (post text, card title, CTA enum), character limits (280 post, 70 card title), asset ratios, and the native-feel rules that decide whether the algorithm classes it as quality content or scroll-past. Use whenever the task is producing X ad copy or card assets; strictly platform-specific, so it does not cover audience targeting, conversion tracking, or organic posting strategy.

Draft and revise Medium stories in the author's documented voice, with sourced proof and a deliberate business purpose, covering publication submissions, native launches, canonical syndication, disclosure, reader access, and measurement. Activate for writing for Medium, publishing on Medium, a Medium draft or piece, or revising a Medium publishing plan.

Specifies what to produce for a LinkedIn ad: Sponsored Content (single image, video, carousel, document), Message Ads, and Text Ads, with the exact slots (intro text, headline, description), character limits, asset ratios, the CTA enum per format, and the professional-context rules that decide whether B2B audiences engage or scroll past. Use whenever the task is producing LinkedIn ad copy or asset briefs; strictly platform-specific, so it does not cover audience filters, matched audiences, or LinkedIn Insight Tag setup.

Operational playbook for diagnosing the off-page link surface of a site: referring-domain quality distribution, unique-root-domain count, anchor-text distribution and over-optimization risk, toxic-neighborhood detection, paid-link risk patterns, link gap vs named competitors, and community/forum presence (Reddit and niche communities, which hold many SERP and AI-answer citations). Encodes the fetch protocol (parallel web search + page reads + competitor backlink discovery), the per-check rubric, the severity matrix, and the unverified-check escalation rule. Use when auditing the link profile of a site or a specific URL. Output: prioritized link-profile findings with cited evidence and fix direction.

Reference for driving the octomind binary: the run, server, acp, login, config, tap, untap, vars, send, evaluate, workflow, and completion commands and their flags, the interactive slash commands, provider API-key environment variables, .env precedence, and OCTOMIND_* overrides. Activate when a user asks how to invoke octomind, which flag or slash command does what, how to resume or daemonize a session, or which environment variable a provider reads.

macOS administration: hardware and release detection, launchd domains, managed updates, security boundaries, packages, diagnostics, and recovery.

Operational playbook for designing production LLM evaluation in 2026. Covers eval framework selection (Promptfoo, DeepEval, Inspect, LangSmith, Braintrust, Phoenix, Galileo), golden dataset construction (size, stratification, versioning, contamination), synthetic data generation, LLM-as-judge design and calibration (G-Eval, human agreement, position/length bias, preference leakage), statistical significance (standard errors, clustering, paired differences, power analysis), RAG metrics (RAGAS, TruLens RAG Triad), agent evaluation (outcome grading, pass@k vs pass^k, transcripts), drift detection, eval-driven development and CI regression gates, and benchmark literacy (saturation and contamination of MMLU, GPQA, SWE-bench Verified; SWE-bench Pro, HLE, tau2-bench, BFCL, GAIA). Use when designing an eval suite, validating a prompt or model change, gating a deploy, or diagnosing production quality drift. Output: eval verdicts with statistical evidence, not vibe judgments.

Translate and localize documents, code comments, UI strings, and structured content while preserving formatting, adapting to domain-specific terminology, and maintaining consistency across large files via batching.

Per-language calibration anchors for detecting AI-slop and translationese in German (de) target text; auto-loads alongside content-locale-humanize when the target language is German.

macOS app best practices for AI coding agents: window and command ownership, AppKit/SwiftUI integration, sandboxed files, and distribution. Auto-activates for Mac app projects.

Resolve reported issues (bug reports, feature requests) the way a project maintainer would: understand the reported behavior, find the upstream resolution when one exists, survey the full surface the change belongs to, implement in the project's own pattern, and verify against the report. Activate when the task is a reported issue against a codebase.

Per-language calibration anchors for detecting AI-slop and translationese in Japanese (ja) target text; auto-loads alongside content-locale-humanize when the target language is Japanese.

Diagnose and maintain Ubuntu systems: LTS support, APT repositories, phased updates, Netplan, package isolation, service persistence, and release upgrades.

Write and revise Mastodon posts, replies, launches, and proof updates in the author's documented voice, covering discovery, visibility, accessible publishing, disclosure, and qualified follow-up. Activate for Mastodon or Fediverse publishing briefs; check the actual server before applying Mastodon mechanics elsewhere.

Per-language calibration anchors for detecting AI-slop and translationese in Spanish (es) target text; auto-loads alongside content-locale-humanize when the target language is Spanish.

Operational playbook for diagnosing the on-page content layer of a page or page set: title and meta description optimization, heading structure, search-intent match, answer placement and AI-search visibility per Google's 2026 guidance, E-E-A-T proof (author byline, experience signals, named specifics), and information-gain vs the top 10 SERP results. Encodes the per-check rubric, the severity matrix, and the unverified-check escalation rule. Use when auditing the on-page content surface of one or more URLs. Output: prioritized content findings with cited evidence and fix direction.

Per-language calibration anchors for detecting AI-slop and translationese in Thai (th) target text; auto-loads alongside content-locale-humanize when the target language is Thai.

Idiomatic Go best practices for AI coding agents: architecture, concurrency design, and standard-library-first patterns. Auto-activates in Go projects.

Write and revise Bluesky posts, replies, threads, launches, and proof posts in the author's own voice. Activate for Bluesky drafting, format selection, community participation, or a publishing package with sourced claims and an explicit next step.

iOS app best practices for AI coding agents: state ownership, SwiftUI/UIKit integration, persistence, lifecycle, and platform services. Auto-activates for iOS projects.

Write YouTube scripts and publishing packages in the author's documented voice, covering format choice, honest packaging, spoken delivery, and qualified next steps. Activate for titles, thumbnails, descriptions, Shorts, channel posts, product launches, demonstrations, customer proof, or revisions using audience feedback.

Zendesk MCP Server - tickets, users, organizations

MCP SQLite Server - query SQLite databases

MCP PostgreSQL Server - query PostgreSQL databases

MCP Slack Server - interact with Slack workspaces (channels, messages, users)

MCP Fetch Server - retrieve web content and convert to markdown

MCP Memory Server - persistent key-value store for context

MCP Puppeteer Server - browser automation and web scraping

MCP Sequential Thinking Server - structured AI reasoning and problem-solving

MCP Git Server - read, search, and manipulate local Git repositories

MCP Filesystem Server - secure file operations (read, write, search)

Exa MCP Server - web search, code search, and company research

Cloudflare MCP Server - Workers, KV, R2, D1, DNS management

Shopify MCP Server - products, orders, customers, inventory

Firecrawl MCP Server - web scraping, crawling, and data extraction

Whisper MCP Server - audio transcription with OpenAI Whisper

Webflow MCP Server - manage Webflow sites, pages, and CMS collections

Ahrefs MCP Server - SEO, backlinks, keywords, rank tracking

Hedra Character-3 MCP Server - image-to-talking-character video

Google Maps MCP Server - geocoding, navigation, location services

Tavily MCP Server - AI search API for real-time web information and extraction

Mobile MCP Server - iOS & Android device automation (real devices + simulators/emulators)

Vercel MCP Server - deployment and project management

Vitest MCP Server - run Vitest tests, list test files, and analyze coverage

MiniMax Hailuo 02 MCP Server - text-to-video and image-to-video generation

Pinecone MCP Server - vector database for AI/ML

Prometheus MCP Server - query Prometheus metrics, targets, and labels

Context7 MCP Server - up-to-date library documentation and code examples

Ansible MCP Server - playbook docs and automation info

Meta Instagram Graph API MCP Server - Reels and feed video publishing

Yahoo Finance MCP Server - stock prices and historical market data

Google Cloud MCP Server - GCP services management

Remotion MCP Server - code-driven (React) video composition + Lambda render

React MCP Server - official React documentation and compiler tools

Twilio MCP Server - SMS, voice, messaging

Notion MCP Server - interact with Notion workspaces (pages, databases, blocks)

Readwise MCP Server - reading highlights, books, and document access

YouTube MCP Server - video subtitles, transcripts, channel info

Azure MCP Server - Azure cloud services management

Amplitude MCP Server - product analytics charts, dashboards, cohorts, and experiments

Discord MCP Server - Discord bot integration, messaging, moderation

HeyGen MCP Server - AI avatar UGC video generation

SE Ranking MCP Server - SEO data, keyword research, site audits

DeepL MCP Server - text and document translation via DeepL API

AssemblyAI MCP Server - diarized captions, chapters, speaker labels, sentiment

Brave Search MCP Server - web search via Brave Search API

GitHub MCP Server - full GitHub API operations (repos, issues, PRs, actions)

Affise MCP Server - affiliate marketing campaigns, analytics

Instagram MCP Server - posts, comments, DMs, stories via Graph API

TailwindCSS MCP Server - utility classes, colors, CSS conversion

CourtListener SDK - US court cases, legal research, and opinions

Replicate MCP Server - image generation with Flux

macOS Notifier MCP Server - send native macOS notifications

Pulumi MCP Server - Infrastructure as Code

ElevenLabs MCP Server - text-to-speech, voice cloning

Trello MCP Server - boards, cards, lists

Freepik MCP Server - AI generation + stock images, vectors, video templates

n8n MCP Server - workflow automation builder

xurl MCP bridge - official X API MCP (api.x.com/mcp) with OAuth handled locally

Sentry MCP Server - error tracking, issues, and performance monitoring

Gmail MCP Server - send, search, read emails and manage labels

Google Veo MCP Server - text-to-video and image-to-video via Vertex AI / Gemini API

YouTube Data API v3 MCP Server - Shorts and long-form upload, metadata edit, analytics

Google Drive MCP Server - search, read, and manage Google Drive files

Google Search Console MCP - search analytics, URL inspection

Google Calendar MCP Server - create events, manage calendars, check availability

E2B MCP Server - secure code execution sandbox for Python/JS

Medical MCP Server - FDA, WHO, PubMed, RxNorm medical data

Calendly MCP Server - scheduling, events, invitees, and availability

Stable Diffusion MCP Server - local image generation and upscaling via SD WebUI

German Law MCP Server - federal statutes live from gesetze-im-internet.de

Obsidian MCP Server - read, create, search notes in Obsidian vault

Luma Dream Machine MCP Server - AI video generation via the RunAPI gateway

MongoDB MCP Server - official MongoDB Atlas and database operations

Supabase MCP Server - Postgres database, auth, and storage

macOS Contacts MCP Server - access the macOS Contacts database

React Aria MCP Server - Adobe's official React Aria documentation

Email MCP Server - SMTP/IMAP email via standard protocols

Intercom MCP Server - customer support tickets

TikTok Shop MCP Server - products, orders, inventory

HubSpot MCP Server - CRM contacts, deals, companies

Terraform MCP Server - Infrastructure as Code

macOS Automator MCP Server - AppleScript, JXA execution

Binance USDT-M Futures MCP Server - market data, orders, positions, margin (requires uvx)

Airtable MCP Server - read and write Airtable bases, tables, and records

Jenkins MCP Server - CI/CD pipeline management

Mubert MCP Server - royalty-safe AI music generation for ad soundtracks

PayPal MCP Server - payments, transactions

Kubernetes MCP Server - manage K8s clusters (pods, deployments, logs, helm)

Linear MCP Server - project management (issues, projects, teams)

Spotify MCP Server - music playback, playlists, library

LlamaCloud MCP Server - LlamaIndex managed indexes

Jira MCP Server - issue tracking, projects, comments, workflows

MCP WhatsApp Server - interact with WhatsApp Web (messages, contacts, media)

Social Media MCP - multi-platform posting, scheduling

MCP Telegram Server - interact with Telegram channels and groups

Asana MCP Server - projects, tasks, workspaces

Sync.so MCP Server - lipsync postprocessing for video

Auth0 MCP Server - manage Auth0 applications, APIs, actions, and logs

Social Media Handle Checker MCP Server - username availability on YouTube, TikTok, Instagram, Threads, X

Svelte MCP Server - official Svelte and SvelteKit docs and tooling, run via bunx

Microsoft Teams MCP Server - send messages, manage chats, and search Teams channels

Playwright MCP Server - browser automation, screenshots, web scraping

Serpstat MCP Server - SERP tracking, backlinks, domain authority

Social Media Scheduler MCP Server - multi-platform content scheduling

Salesforce MCP Server - CRM objects, records, queries

OpenAI Assistants MCP Server - use the OpenAI Assistants API from AI tools

ChatGPT MCP Server - send queries to ChatGPT from other AI tools

DALL-E MCP Server - OpenAI DALL-E image generation

OpenAI Sora MCP Server - video generation via the OpenAI Video API

Kling AI MCP Server - text-to-video and image-to-video generation

Figma MCP Server - design-to-code integration, read Figma files

Mem0 MCP Server - persistent memory storage and retrieval for AI agents

UK Legislation MCP Server - official legislation.gov.uk API

Stripe MCP Server - payments, customers, subscriptions, invoices

Redis MCP Server - key-value store operations via the reference MCP server

Hugging Face MCP Server - ML models, datasets, Spaces discovery

AWS S3 MCP Server - object storage, upload, download, list buckets

GitLab MCP Server - repos, merge requests, pipelines, and issues

UK Law MCP Server - access UK legislation, statutes, and legal references

Canadian Law MCP Server - access Canadian legislation, statutes, and legal references

French Law MCP Server - access French legislation, statutes, and legal references

German Law MCP Server - access German legislation, statutes, and legal references

macOS System MCP Server - file dialogs, clipboard, notifications

Qdrant MCP Server - vector search engine for semantic search and retrieval

Open-WebSearch MCP Server - keyless multi-engine web search (DuckDuckGo, Bing, Startpage, Brave)

TikTok Content Posting API MCP Server - direct upload of videos and photos

Desktop Automation MCP Server - mouse, keyboard, screenshots

Grafana MCP Server - dashboards and data visualization

Refersion MCP Server - influencer, affiliate marketing tracking

Datadog MCP Server - monitoring and observability

Pika Labs MCP Server - text-to-video, image-to-video, Pikaffects

AppleScript MCP Server - macOS automation and control via AppleScript

Pexels MCP Server - free b-roll video and photo stock

Resend MCP Server - transactional email sending

RunwayML MCP Server - Gen-4 video/image generation

ffmpeg - multimedia framework for stitching, transcoding, and burning captions

Installs uv - fast Python package and project manager (used by uvx-based MCP servers)

Ensures the Swift toolchain (swift, swiftc) is available - Xcode CLT on macOS, swiftly on Linux, winget on Windows

Installs Rust toolchain via rustup (provides cargo, rustc)

ripgrep (rg) - fast recursive regex search tool for code and text

Installs Docker CLI and daemon (required for containerized MCP servers)

ast-grep (sg) - structural code search and rewrite tool

Installs the octofs CLI (file editing and viewing tool by Muvon)

Installs the octocode CLI (code indexing and search tool by Muvon)

Installs the octobrain CLI (memory and knowledge tool by Muvon)

Installs Node.js and npx for running npx-based MCP servers

Installs Bun - fast all-in-one JS/TS runtime, package manager, bundler, and test runner

Triages a broken thing at home, in the car, or with a pet; stops immediately on danger signs, otherwise routes to a specialist for a ranked diagnosis, DIY-vs-pro call, and fix plan checked by an independent safety auditor.

Turns a bank or card statement export into a subscription and fee audit with every figure computed from the transactions, independently recomputed by an auditor loop, and ranked by annualized savings.

Transcreates content into every requested language, working each one in its destination file in-place (or inline when none is given), loops fix and audit until each passes native-fluency review, then ends with a per-language summary.

Runs a code review of the current unstaged changes against the request, independently verifies each finding, then branches on a deterministic verdict to an approval note or a prioritized fix list.

Upgrades dependencies (named packages, or safe patch/minor by default), fixes breakages in a bump/verify loop until the project's checks pass, then writes upgrade notes.

Security-audits the current repo across OWASP lenses, then fixes and independently re-audits in a loop until findings are resolved and checks stay green.

Turns a data source into a decision-ready data analysis report with every number computed from the data and inline-SVG charts, refined by an accuracy loop, delivered as self-contained HTML that prints to PDF.

Grounds project memory from the codebase, promotes durable knowledge into the .box/ knowledge base, then audits and fixes the notes until they pass an independent audit against the repo.

Takes a product idea from market exploration to an honest go/no-go, then, only if viable, produces a go-to-market plan with GTM strategy, brand, pitch, ads, and a zero-budget launch plan.

Runs an SEO audit of a site or page across technical, on-page, off-page, and GEO (generative engine optimization) lenses, then turns the evidence-bound findings into a tiered, finding-traceable, KPI-tagged SEO strategy brief.

Turns an existing article into grounded, platform-native social media drafts for each network requested, polished by a fix/audit loop until all pass, then delivered to files or inline. Chain after content.

Writes a blameless incident postmortem report from pasted evidence: builds a timestamped timeline, analyzes contributing factors and response metrics in parallel, and loops an independent judge until it is traceable, blameless, and arithmetically correct.

Turns a trip request into a day-by-day travel itinerary and budget table via parallel logistics, lodging, and activities research, then loops an independent feasibility audit on pacing, transit, opening hours, and budget, ending with a verify-before-you-book list.

Writes behavior tests for the changed code or a named module, then proves they catch bugs with mutation testing: passing on current code, touching only test files, and failing under deliberate mutations that are reverted exactly.

Turns a meeting transcript or notes into meeting minutes, decisions, owner-and-date action items, and follow-up drafts, each traced to a verbatim quote by an independent verifier loop that flags missing owners and dates instead of inventing them.

Spec-driven development loop for an existing feature spec: gathers context, then iterates a developer/evaluator loop whose independent evaluator re-runs the build/test gate, proves every requirement's test fails on the old code and passes on the new, and rejects tampered tests until it approves, then summarizes the change.

Builds a weekly meal plan and grocery list from household constraints: a 7-day meal plan with recipes, aisle-grouped grocery list, and prep schedule, then loops an independent auditor on allergens, list coverage, daily time, and budget until it passes.

Checks an issue report against the current repo, verifies every code anchor, then produces a CONFIRMED / NEEDS_INFO / NOT_A_BUG verdict with labels, priority, and a maintainer reply draft; never posts.

Classifies the current changes into a SemVer bump and change buckets, drafts README, changelog, and release notes in parallel, then reconciles them into one consistent plan via a validating gate.

Tailors your resume to a job description: turns a master resume and a job posting into an ATS-optimized resume, a cover letter, and an honest match-gap report, refined by a screen/fix loop. Produces documents; never auto-submits.

Turns a master resume and a job posting into a grounded interview prep pack — likely interview questions, STAR answers from real experience, honest gap answers, and questions to ask — refined by a screen/fix loop. Never contacts anyone.

Routes a confusing letter, bill, contract, lease, or notice to the right specialist, composes a plain-language decode with red flags, deadlines, next steps, and a draft reply when warranted, and verifies every figure and date against the source.

Runs a code review of a change from five independent lenses in parallel, adversarially verifies every finding against the real code, then synthesizes one severity-ordered review with a verdict.

Reproduces a bug and locates its root cause in the current repo, iterates a fix/verify loop until a test proves it fixed, then summarizes the change.

Investigates a question from background, evidence, and counter-argument angles in parallel, synthesizes a cited report, then loops a groundedness judge until every claim verifies against its sources.

Extracts structured health facts and gates on red flags first, then analyzes lab results, medications, and symptom context in parallel and delivers a judge-checked, non-diagnostic doctor visit prep brief with ranked questions for your clinician.

Mines a field for evidenced pain via parallel community, review, and demand sweeps, vets candidates adversarially into ranked briefs or an honest no-signal verdict. Chain the winner into launch.

Turns a feature request into an EARS behavioral spec with testable requirements, stops with clarifying questions when anything is unresolved, then implements it in a developer/evaluator loop whose independent evaluator re-runs the checks, proves every requirement's test fails on the old code and passes on the new, and rejects tampered tests before summarizing the change.

Turns a brief into a researched, publish-ready article in the current directory, refined by an edit/audit loop until it passes. Chain the promote workflow for social adaptation.