workflow
URL mode
Extract design tokens/cues from a URL.
src/modes/url.ts
All 363 indexed internal units: skills, agents, commands, modules, workflows, plugins, MCP servers, CLIs, prompts, templates, and other routable pieces. Each component page preserves triggers, anti-triggers, inputs, outputs, dependencies, services, activation, compatibility, limitations, security notes, evidence, and related components.
Extract design tokens/cues from a URL.
Use browser-backed inspection/screenshots.
Analyze local code.
Analyze repository source.
Write extracted knowledge as an agent skill.
Structured multi-agent review.
Explore, architect, implement, and review features.
Specialist PR review agents.
PreToolUse security guidance hook.
Create/validate plugins, agents, skills, hooks, commands, MCP.
Repeated-agent loop with cancel/stop controls.
DOCX/PDF/PPTX/XLSX artifact workflows.
Frontend design guidance.
MCP server/tool design guidance.
Create/improve Agent Skills.
Browser-based webapp testing.
Generate brand-neutral short.
Generate Archon-branded short.
Generate Anthropic-themed short.
Research/script/compose the short.
Generate narration and timing.
Sync/async HTTP fetching.
Anti-bot-oriented fetching.
JS/browser-backed fetching.
Concurrent crawl/session/proxy workflows.
Expose scraping tools to agents.
Route requests among providers/models.
Off/lite/standard/aggressive/ultra/RTK/stacked modes.
Configure/test compression engines and rules.
Expose gateway/cache/compression tools.
Agent-to-agent JSON-RPC.
Visual gateway control plane.
Search web and return LLM-ready results.
Extract page to Markdown/JSON/screenshots.
Discover/scrape multi-page sites.
Browser click/scroll/write/wait actions.
Process URL sets.
General frontend design/audit guidance.
Translate visual references to frontend.
Guide redesign of incumbent UI.
High-end/minimal/brutalist visual guidance.
Reference-board/brand-kit workflows.
Build/execute node workflows.
Compose models/tools/memory/human steps.
Connect external apps/APIs.
Reusable workflow examples.
Turn requirements into a design-system proposal.
Search visual styles.
Select palettes.
Select font pairs/rules.
Apply UX and stack rules.
Create/preview prototypes/decks/images/video.
Persist reusable design context.
Repeat agent-driven design workflows.
Extend capabilities.
Connect supported agent CLIs/MCP.
Configure/check integrations.
Teach agents how to route web access.
Route platform-specific tools.
Optional Playwright/browser-cookie support.
Route frontend requests/playbooks.
Shape/init/document/extract.
Critique/audit.
Polish/harden/animate/layout/adapt/optimize and more.
Browser visual variants.
Deterministic design checks.
Relational data/extensions.
Authentication/authorization.
REST/GraphQL over Postgres.
Realtime events/DB changes.
Object/file storage.
Admin dashboard.
Deploy/manage Vercel projects.
Framework/build/deployment utilities.
Render live status.
Configure statusline/presets.
Drive extraction/generation.
Parse supported formats.
Validate generated skill.
Compare token access costs.
Initialize/manage integrations/templates.
Define project principles.
Define what/why.
Create technical plan.
Generate actionable tasks.
Execute approved tasks.
Improve artifact quality/consistency.
Point to active successor.
Route work through sized delivery workflow.
Route to next/optional workflow.
Product/architecture perspectives.
UX/dev/test perspectives.
Route to a suitable skill.
Elicit requirements/domain language.
Red-green-refactor.
Feedback-driven bug diagnosis.
Standards/spec review.
Convert intent to executable work.
Survey architecture opportunities.
Top-level autonomous build entry.
PM/architecture/lead planning.
Implement and validate.
Integrate and verify acceptance.
Adapt after failures.
Operate Arcads generation endpoints.
Build/clone static ads.
Publish paused Meta ads.
Storyboard/image-to-video/stitching.
Transcribe/burn captions.
Generate thumbnails.
Capture/inject observations.
Process retrieval/storage APIs.
Persist/search observations.
Semantic retrieval.
Search history on demand.
Inspect memory stream.
Code-task examples.
Copy/social/business examples.
Analysis/tutoring examples.
Curate docs/learning.
Curate clients, infra, security, orchestration.
Curate reusable context/tooling.
Index MCP server projects/resources.
Index/search code.
Expose search to MCP clients.
IDE semantic search.
Manage Claude subprocesses/sessions.
PreToolUse approve/deny UI.
Fetch/install skills.
Desktop UI.
Controlled IPC.
Four coding principles.
Same guidelines in Cursor.
Install guidelines globally.
Browse/resume projects/sessions.
Create/run custom background agents.
Token/cost analytics.
Configure/test MCP.
Restore/fork/diff sessions.
Edit project instructions.
Read-only Codex review.
Challenge assumptions/risks.
Delegate investigation/fix.
Run delegated Codex task.
Import Claude session into Codex.
Manage background Codex jobs.
Index skills by domain.
Connect agents to apps through Composio.
MCP Builder/Skill Creator/Webapp Testing examples.
Explore requirements.
Create implementation plan.
Execute plan.
Refine code before review.
Multi-agent report-only review.
Persist solved-problem knowledge.
Bounded autonomous pipeline through PR/CI, no merge.
Create/update isolated profile/runtime.
Local CRM UI.
Dedicated gateway/profile.
Fast 4B generation/editing; Apache-2.0 model variants.
Higher-capacity non-commercial variants.
32B high-quality research/tuning model.
Interactive generation/editing.
Expand prompts locally/OpenRouter.
Parse events/predict spend.
Optional Gemini coach.
Dashboard/calendar/challenges/coach.
Publish static frontend.
Assess, roadmap, optionally build next step.
Validate/run JS handlers in micro-VM.
Path-jailed file/domain-scoped HTTP capabilities.
PPTX/XLSX/PDF/Markdown/HTML outputs.
Expose approved MCP tools as host modules.
pptx/pdf/xlsx/report/data/API/web/MCP guidance.
Fixed data/tokenizer/eval utilities.
Only file agent modifies.
Human-authored research-agent policy.
GPT/optimizer/data/tokenizer/eval/inference.
Train/eval/SFT/RL/chat/tokenizer/benchmarks.
Speedrun/scaling/miniseries/CPU scripts.
Evaluation task adapters.
Feature/test server.
Web fetch/conversion.
Controlled file operations.
Git repository tools.
Knowledge-graph memory.
Reflective problem-solving tool.
Time/timezone conversion.
Issues/agents/projects/review/audit UI.
API/task orchestration/persistence.
Spawn supported agent CLIs on controlled machines.
Reusable solved-problem playbooks.
Scheduled standups/audits/reports.
Organize people/agents with routing leader.
LLM-assisted tree generation.
Fast heuristic tree extraction.
Tree-search retrieval demo.
Vectorless/vision RAG notebooks.
Pack local/remote repos.
Estimate context size.
Extract structural code with --compress.
Exclude known credential patterns.
Browser-based packing.
Filter/group/truncate/dedupe command output.
Rewrite shell commands through RTK.
Failure-focused test/build output.
Compact git/gh.
Compact AWS/Docker/K8s/IaC.
Refine/validate design.
Isolated workspace/baseline.
Detailed implementation plan.
Fresh agent per task + reviews.
Red-green-refactor.
Plan-based review.
Verify and merge/PR/keep/discard choices.
API/backend/frontend/fullstack/mobile/UI agents.
Language/framework experts.
Cloud/DevOps/SRE/IaC agents.
Testing/review/security/compliance agents.
Data/ML/LLM agents.
Install/coordinate other agents.
Create and edit Obsidian Flavored Markdown with wikilinks, embeds, callouts, properties, and other Obsidian syntax.
Create and edit Obsidian Bases files with views, filters, formulas, properties, and summaries.
Create and edit JSON Canvas files containing nodes, edges, groups, and spatial relationships.
Operate an Obsidian vault through the Obsidian CLI, including supported plugin and theme-development tasks.
Extract clean, readable Markdown from a web page with Defuddle while removing navigation and other clutter that wastes context.
Instruct a compatible agent to answer concisely while preserving code, commands, and error text.
Route a supported coding agent through a local proxy that applies content-aware request compression.
Read local agent history without editing it, rank token sinks, and show a report with suggested fixes.
Run a noisy command through recoverable output compression with the caveman shrink command.
Expose a focused, compressed representation of a locally controlled Chrome page.
Remember and recall compact facts while retaining a recovery path for original content.
Use a read-only subagent to locate relevant code with compact path-and-line evidence.
Re-encode suitable structured tool results into a smaller representation and decode them when needed.
Render exceptionally dense text slabs as images when the measured token gate predicts a benefit.
Expose compression, retrieval, statistics, and TOON operations to an MCP-compatible host.
Index a repository into the persistent graph and hand subsequent freshness to auto-sync.
List indexed projects and their current node and edge counts.
Remove a selected project and its graph data after an explicit request.
Report whether a project is indexing, ready, stale, or failed.
Search graph nodes with label, name, file, degree, pagination, and other structural filters.
Traverse incoming or outgoing call relationships to explain who calls a symbol and what it calls.
Map a Git diff to affected symbols, blast radius, and a risk classification.
Run a supported read-only Cypher-like query against the project graph.
Describe available labels, properties, relationship patterns, and counts before custom queries are written.
Return source for a function or method identified by its qualified name.
Summarize languages, packages, routes, hotspots, clusters, and architectural decisions.
Run grep-like text search over files in an indexed project.
Create, read, update, and organize Architecture Decision Records under the project mutation guard.
Ingest runtime traces that can validate or refine inferred HTTP-call relationships.
Explore the local repository graph visually through the built-in localhost interface.
Review prose against 35 documented AI-writing patterns and revise it for naturalness, specificity, rhythm, and voice while protecting meaning and quotations.
Plan statistically sound experiments, define hypotheses and success metrics, and document how results will be interpreted.
Develop and review paid-ad concepts, hooks, copy variants, formats, and creative iteration plans.
Plan and improve paid campaigns across major ad platforms, including targeting, tracking, creative, and optimization.
Improve the chance that content is cited or recommended in AI-generated search answers while preserving conventional SEO fundamentals.
Design, audit, and implement trustworthy analytics events, properties, dashboards, and measurement plans.
Audit and optimize App Store and Google Play listings, keywords, creatives, conversion, and reporting.
Choose and interpret marketing attribution models, connect spend to outcomes, and explain uncertainty in channel credit.
Reduce voluntary and involuntary churn with cancellation flows, save offers, dunning, and retention measurement.
Identify partners and structure joint campaigns, content, events, distribution, and reciprocal value.
Write concise B2B cold-email and follow-up sequences built around relevance, credibility, and a clear next step.
Plan, launch, and grow an online community that creates durable member value and supports the product.
Research a competitor from primary evidence and produce a structured profile of positioning, audience, product, and go-to-market signals.
Create honest competitor-comparison and alternative pages for search discovery and sales enablement.
Choose audiences, themes, formats, distribution, and an editorial system for a coherent content program.
Diagnose and improve existing marketing copy for clarity, specificity, hierarchy, credibility, and conversion intent.
Draft or rewrite marketing copy for homepages, landing pages, product pages, campaigns, and other conversion surfaces.
Audit and improve page or form conversion using evidence, friction analysis, prioritization, and testable recommendations.
Plan interviews and surveys, synthesize customer evidence, and turn findings into useful language and decisions.
Build and execute a controlled submission plan for relevant product, startup, SaaS, AI, and review directories.
Design lifecycle, onboarding, nurture, launch, and re-engagement email sequences with clear goals and measurement.
Evaluate and plan useful free tools that can earn attention, search demand, leads, or product-qualified traffic.
Plan, create, edit, and optimize marketing imagery for web, product, social, advertising, and editorial use.
Find suitable creators and structure outreach, briefs, agreements, tracking, and campaign evaluation.
Build a phased launch or announcement plan covering audience, narrative, channels, assets, timing, and follow-through.
Design a genuinely useful downloadable or interactive asset that earns qualified email signups.
Examine a marketing decision through several named expert perspectives, then synthesize disagreements and recommendations.
Generate and prioritize practical marketing ideas that match a product, audience, constraints, and growth stage.
Design repeatable agent-assisted marketing routines with triggers, inputs, review gates, outputs, and feedback.
Create an integrated marketing plan with research, positioning, channels, campaigns, metrics, owners, and sequencing.
Apply behavioral-science principles carefully to messaging, choice architecture, motivation, and conversion design.
Strengthen what is being sold through value framing, packaging, bonuses, guarantees, proof, and risk reversal.
Improve post-signup activation and time-to-value through journey mapping, education, prompts, and instrumentation.
Plan and optimize upgrade screens, feature gates, upsells, and in-product monetization moments.
Create and evaluate overlays, banners, slide-ins, and modals without sacrificing relevance or usability.
Research and reason about pricing, packaging, value metrics, tiers, willingness to pay, and rollout risk.
Create and maintain the shared product-marketing context that downstream skills use for audience, positioning, and proof.
Plan scalable, data-backed search pages with useful templates, quality controls, internal linking, and indexation safeguards.
Define an ideal customer profile and build a qualified, evidence-backed prospect list for responsible outreach.
Develop media angles, press materials, journalist outreach, and an earned-media plan grounded in real news value.
Design or improve referral, affiliate, and word-of-mouth programs with aligned incentives and fraud controls.
Map lead and revenue operations, lifecycle stages, routing, handoffs, data quality, and reporting responsibilities.
Create practical collateral, talk tracks, objection responses, one-pagers, demo plans, and buyer-facing materials.
Select, implement, and validate appropriate structured data without marking up claims the page does not support.
Run a structured technical, on-page, content, and authority review and prioritize issues by likely impact and effort.
Reduce avoidable friction in registration, account creation, verification, and trial-start flows.
Plan page hierarchy, navigation, URLs, and internal links so people and crawlers can understand the site.
Plan compliant, useful SMS or MMS journeys with consent, timing, frequency, copy, and measurement guidance.
Create channel-aware social content, calendars, repurposing systems, and performance feedback loops.
Plan and produce marketing video concepts, scripts, assets, workflows, and platform-specific deliverables.
Discover toolkits, create user-scoped sessions, retrieve tools, and execute actions from TypeScript.
Use the same core API with a smaller package that omits bundled source and documentation.
Use Composio sessions, tools, authentication, and execution from Python applications.
Search, link, execute, and script toolkit operations from a terminal or coding agent.
Translate Composio tools into the native tool format of supported agent frameworks.
Find a small relevant subset of actions from the hosted catalog instead of loading every available schema.
Link and manage user-specific OAuth or API-key connections for external services.
Subscribe to supported external events and deliver them to an agent application.
Give an agent a controlled environment for files, scripts, and longer-running tool workflows.
Produce a general animated video through research, script, asset, scene, composition, edit, review, and publish stages.
Produce presenter-led video using an avatar or generated spokesperson with coordinated voice, scenes, and publishing output.
Design, rig, animate, edit, and review a character-centered production.
Plan and assemble a cinematic sequence with research, visual direction, generated assets, edit decisions, and final review.
Generate a repeatable set of short clips from a defined idea, source, or campaign system.
Build a sourced documentary-style montage from research, narration, real or generated footage, and editorial structure.
Turn a concept into a structured explanatory video with script, scenes, visual teaching aids, and quality gates.
Combine multiple production techniques or provider types under one coordinated project workflow.
Translate, adapt, voice, synchronize, and review an existing video for another language or market.
Transform long-form podcast material into selected clips, scenes, captions, and platform-ready outputs.
Create a product or workflow demonstration from screen recordings, scripted narration, edits, and callouts.
Produce a presenter-centered video with script, voice/performance direction, supporting assets, edit, and publishing stages.
Remove low-information AI mannerisms and noisy boilerplate while preserving the intended technical result.
Send a bounded question to a configured external provider and bring the response back into the active workflow.
Run a staged plan-build-verify workflow from a high-level goal with checkpoints and recoverable state.
Execute an iterative, evidence-recorded research loop for measurable technical experiments.
Stop an active OMC mode or workflow and clean up its tracked state safely.
Ask multiple coding providers for complementary analysis and have Claude synthesize their findings.
Configure desktop or external notifications for long-running OMC activity.
Drive a reproduction-first debugging workflow that traces symptoms to root cause and verifies the fix.
Perform a deliberate, high-context investigation of a codebase, architecture, or difficult technical question.
Clarify an ambiguous project through structured questions before planning or implementation begins.
Create a deeper project initialization and context map for subsequent agent work.
Gather explicitly requested outside context and package it for the active Claude Code session.
Display compact workflow, context, and agent-status information during a session.
Capture reusable lessons from completed work and convert them into future guidance.
Keep source-build users aware when generated or compiled artifacts must be refreshed.
Configure supported MCP integrations and verify that Claude Code can reach them.
Assess tests, review findings, conflicts, and delivery evidence before a branch or pull request is merged.
Diagnose installation, configuration, runtime, hook, and dependency problems in OMC.
Open the consolidated OMC command, mode, configuration, and compatibility reference.
Install and configure OMC's project and user-level files through a guided setup flow.
Compatibility entry point that directs work to the current team orchestration surface.
Turn a request into an ordered implementation plan with explicit scope and verification steps.
Create and manage project sessions, worktrees, templates, and provider-specific execution contexts.
Persist on an implementation goal through bounded retries and verification until its completion condition is met.
Iterate on a plan with critique and refinement before expensive execution begins.
Coordinate versioning, validation, changelog, packaging, and release-readiness work.
Save an explicit fact or decision to OMC's durable project memory.
Apply a scientific hypothesis-experiment-measurement loop to technical investigation.
Evaluate agent behavior against goals and propose measured configuration or workflow improvements.
Provide the short in-session entry point to OMC installation and configuration.
Inspect, select, and use an installed OMC skill deliberately.
Convert a repeatable workflow or source document into a structured reusable skill.
Coordinate Claude Code's native multi-agent team through planning, execution, verification, and repair stages.
Trace agent decisions, hooks, state transitions, or execution paths for diagnosis.
Manage a large goal with durable state, decomposition, progress checks, and completion criteria.
Run a deeper quality-assurance pass across behavior, tests, edge cases, and deliverables.
Use higher-effort parallel execution for a broad, well-scoped implementation workload.
Check claims and deliverables against concrete tests, commands, artifacts, or acceptance criteria.
Evaluate visual output against explicit design and usability criteria.
Build and query durable project documentation from repository evidence.
Maintain writing-specific preferences, facts, terminology, and continuity across sessions.