OSOP Hub

Lint, review, and approve code changes with AI + human oversight.

Human provides brief, AI generates draft, human reviews and iterates.

Upload data, AI analyzes trends, human confirms before distribution.

AI scores and ranks candidates, HR makes final interview decisions.

AI translates, native speakers review, publish to multiple locales.

Benchmark multiple models, compare metrics, generate recommendation.

AI scans contracts for risk clauses, lawyer reviews flagged items.

AI analyzes patient data and suggests diagnoses; physician makes final call.

AI analyzes customer browsing and purchase history to generate personalized recommendations.

AI analyzes sales data and food costs to recommend menu changes.

AI runs comparable analysis and market adjustments; appraiser validates.

AI assesses damage from photos and checks coverage; adjuster makes final decision.

AI plans personalized itinerary, checks availability, traveler approves before booking.

AI analyzes network topology and predicts congestion; engineer approves config changes.

AI forecasts energy demand from historical data; operator validates dispatch plan.

AI analyzes soil and weather to recommend crops; farmer reviews before planting.

AI checks structural integrity and code compliance; engineer approves.

AI generates personalized workout plan from health data; trainer reviews and delivers.

AI designs curriculum from learning objectives; instructor reviews before publishing.

AI extracts clauses and compares against templates; legal team annotates and returns.

AI extracts entities and builds knowledge graph from documents; expert validates.

AI detects defects from product images and classifies severity; QA inspector decides.

Structured incident handling with AI triage and human oversight.

System monitors metrics, AI classifies severity, routes to human.

AI auto-reviews user content; flagged items go to human moderators.

Real-time transaction scoring with human review for edge cases.

AI monitors patient vitals, alerts clinical team on deterioration.

AI monitors inventory levels, predicts demand, and alerts purchasing managers.

AI analyzes sensor data to predict equipment failures before they occur.

AI scans upcoming lease expirations, compares market rates, and alerts property managers.

AI pattern-matches incoming claims against known fraud indicators.

AI detects unusual booking patterns and alerts revenue managers for pricing adjustments.

AI detects degradation patterns in network telemetry and alerts NOC before outages.

AI forecasts electricity demand peaks and alerts grid operators for demand response.

Drone imagery analyzed by AI to detect pest infestations and alert agronomists.

AI analyzes site camera feeds to detect PPE violations and fall risks in real time.

AI monitors service metrics, predicts SLA breaches, and alerts ops for preemptive action.

AI identifies recurring payment failure patterns and alerts finance for root cause resolution.

AI scores customer engagement and alerts CSMs before high-value accounts churn.

IoT sensors monitor storage conditions; AI flags violations for kitchen managers.

Wearable data analyzed by AI to detect abnormal patterns and alert fitness trainers.

Coordinator spawns 3 parallel explore agents, then synthesizes findings.

Embed query, retrieve context, generate answer, validate grounding.

Classifier routes to specialized agents by intent, then summarizes.

Try fast model first; escalate to larger model if confidence is low.

Multiple AI analysts argue positions, system aggregates consensus.

CI fails, AI diagnoses and fixes, CI re-runs, human merges.

Competitor scraper feeds demand prediction, elasticity modeling, and pricing engine.

Sensor data flows through anomaly detection, root cause analysis, and maintenance scheduling.

Batch orders, compute distances, solve constraints, assign routes, dispatch.

Application flows through risk scoring, actuarial modeling, and pricing with human review for edge cases.

Grid monitor triggers load forecasting, price signaling, consumer scheduling, and grid verification.

Soil analysis, weather prediction, crop modeling, and irrigation scheduling for optimal yield.

Parallel extraction from two sources, matching, discrepancy classification, and resolution.

NLP intake, sentiment analysis, category classification, priority scoring, and routing.

Demand forecasting, competitor monitoring, yield management, and rate publishing.

OCR extraction, entity recognition, relationship mapping, and knowledge graph construction.

Fault detection, topology analysis, remediation planning, config push, and verification.

Extract BOM, query supplier pricing, find alternatives, optimize cost, produce final BOM.

Health assessment, dietary analysis, meal planning, grocery list generation, and weekly plan delivery.

Three LLMs grade in parallel, scores are aggregated, bias is checked, final grade is issued.

End-to-end new hire onboarding from HR record creation through parallel provisioning to training completion.

Multi-level document approval with manager, legal, and executive sign-off plus revision loops.

Full B2B sales cycle from lead capture through qualification, demo, proposal, negotiation, and close.

Bug lifecycle from report through triage, assignment, fix, verification, and close.

Full agile sprint from backlog grooming through planning, daily standups, review, and retrospective.

Annual performance review with self-assessment, peer feedback, manager evaluation, and calibration.

Scholarly paper submission through editorial screening, parallel peer review, and revision cycles to publication.

Employee submits time-off request through balance check, manager approval, and calendar update.

Post-sale customer onboarding from contract signature through account setup, training, and first value milestone.

Customer-initiated return or exchange from request through inspection, processing, and replacement shipment.

Real estate closing process from accepted offer through inspection, appraisal, title search, and key transfer.

Insurance claim lifecycle from filing through documentation, investigation, settlement negotiation, and payment.

Hotel guest journey from reservation confirmation through pre-arrival, check-in, service, and checkout with feedback.

Patient appointment scheduling with availability check, insurance verification, reminders, and check-in.

Customer complaint handling from intake through investigation, resolution proposal, and follow-up.

Full contract lifecycle from drafting through internal review, counterparty negotiation, redlining, and execution.

Asset lifecycle from request through procurement approval, purchase, tagging, assignment, and periodic audit.

Employee certification workflow from enrollment through coursework, examination, supervisor sign-off, and HRIS recording.

Customer plan change workflow covering eligibility, billing proration, access updates, and confirmation.

Guest dining experience from booking through confirmation, seating, service, payment, and feedback.

Building permit process from plan submission through parallel department reviews, final approval, and permit issuance.

New member lifecycle from signup through health assessment, orientation, program assignment, check-ins, and renewal.

Partner applies → review → issue keys → sandbox test → go live.

RFP → vendor responses → evaluation → background check → contract.

Order → inventory check → ship → track → deliver → confirm.

Transform → validate → transmit → acknowledge.

Receive invoice → OCR → match PO → approve → pay.

Apply → document check → department review → public comment → issue.

Seller applies → compliance check → catalog upload → quality review → activate → monitor.

Buyer PO → capacity confirmation → production → QC inspection → ship → acceptance.

Agent creates listing → MLS submission → portal syndication → lead routing → showing scheduling.

Cedant prepares risk → broker markets → reinsurers quote → negotiate → bind → document.

Search GDS → check availability → compare fares → book → ticket → confirm.

Interconnect request → feasibility → capacity planning → SLA → provision → test.

Generator offers capacity → grid evaluates → bidding → dispatch → metering → settlement.

Farmer lists harvest → certification → buyer bidding → contract → logistics → delivery.

GC posts scope → subs bid → evaluate → award → milestones → payment.

Merchant applies → KYC/AML → integration → certification → go-live → monitoring.

Inquiry → qualification → agreement → site selection → training → grand opening.

Define terms → monitoring → periodic review → breach escalation → remediation → renewal.

Extract records → AI matching → discrepancy report → joint resolution → final reconciliation.

Gym registers → API setup → class sync → member exchange → billing → performance review.

Extract, transform, validate, and load with retry handling.

Monitor load → scale decision → provision → health check → rebalance.

Scheduled backup → verify → replicate → test restore.

Monitor expiry → renew → validate → deploy → verify.

POS transaction triggers inventory update, warehouse sync, and auto-replenishment.

PLC sensor data flows through SCADA to historian DB with alarm evaluation.

Subscribe to MLS feed, normalize, deduplicate, and push to portals.

Claim event triggers policy lookup, rules engine, benefit calc, and payment.

OTA booking event propagates to PMS, channel manager, and analytics.

SNMP polling through metric aggregation, threshold evaluation, and ticketing.

Meter readings drive load aggregation, frequency regulation, and billing.

IoT sensor data from edge to cloud with anomaly detection and irrigation control.

BIM model update triggers clash detection, issue logging, and reporting.

Batch transaction aggregation through clearing, ledger update, and reconciliation.

Usage metering, invoice generation, payment processing, and revenue recognition.

Order placed at POS flows to kitchen display, prep tracking, and payment close.

RFID/barcode scan triggers location update, condition check, and CMDB sync.

Wearable device sync through normalization, metrics, trends, and reporting.

Demonstrates v1.1 foreach iteration — send personalized emails to a user list.

Demonstrates v1.1 switch/case — route customer queries to specialized handlers.

Demonstrates v1.1 sub-workflows — compose complex pipelines from reusable building blocks.

Demonstrates v1.1 join modes — wait_all, wait_any, wait_n for parallel branches.

Demonstrates v1.1 workflow-level timeout — entire pipeline has a 30-minute deadline.

Claude Code debugs and fixes a token refresh race condition using sub-agent exploration and coordinated fix.

High-risk AI workflow for AI-assisted medical diagnosis. Includes mandatory human oversight, risk assessment, and approval gates as required by EU AI Act Article 19 for high-risk AI systems.

Workflow for engaging a new community/ecosystem with OSOP. Used for tracking international adoption waves.

Standard workflow for submitting OSOP adoption PRs to external repositories. Dogfooding: this workflow itself is an OSOP file.

Release workflow for publishing OSOP packages to npm/PyPI. Includes version bump, tests, build, and publish.

Content creation from ideation to publishing.

Collect and act on customer feedback.

Design iteration and approval.

End-to-end recruiting workflow.

Structured incident response.

Documentation lifecycle.

Structured meeting documentation.

Quarterly OKR setting and tracking.

Coordinated product launch workflow.

Product development from idea to shipping.

Demonstrates v1.1 join modes — wait_all, wait_any, wait_n for parallel branches.

Demonstrates v1.1 sub-workflows — compose complex pipelines from reusable building blocks.

Demonstrates v1.1 switch/case — route customer queries to specialized handlers.

Get PR details, check CI status, create deployment, wait for status, post comment.

Complete payment sequence: create customer, payment intent, confirm, check status, send receipt.

First wave of OSOP adoption PRs targeting awesome lists and platform plugin registries. Focus on low-friction, high-visibility targets.

Massive PR wave to top AI frameworks: Dify (55k), MetaGPT (45k), Flowise (35k), AutoGen (35k), CrewAI (25k), Semantic Kernel (23k), OpenAI Swarm (18k), Haystack (18k), Prefect (17k), Phidata (15k), pl

Second wave targeting workflow orchestrators (CAMEL-AI, Kestra, Windmill, Trigger.dev) and more awesome lists (awesome-langchain, awesome-n8n, awesome-argo). CAMEL-AI PR was closed by maintainer ("not

AI reviews code, routes to fast-track or deep-review based on risk score, with fallback on failure.

A 3-node workflow: generate an idea, critique it, then refine it.

Create a new forum post with optional images and tags.

Add a comment to a forum post, optionally attaching multiple images.

Post a nested reply to an existing comment.

Update the title, content, tags, or cover image of an existing post.

Update an existing comment's text and optionally add images.

Delete a forum post by ID.

Delete a comment by ID.

Register a new clinic with profile details and photos.

Modify an existing clinic's details and optionally add images.

Write a star-rated review for a clinic with optional photos.

Edit an existing clinic review, add or remove images.

Write a multi-dimensional treatment review and upload before/after comparison photos.

Register a new doctor with specialties, credentials, and avatar.

Define a new treatment type with pricing, recovery info, and keywords.

Create a professional article authored by a doctor, with optional deferred publishing.

Start a treatment recovery diary and add daily progress entries with photos and health metrics.

Submit a consultation request for a clinic and track its status.

Request ownership of a clinic listing, pending admin verification.

User reports inappropriate content; admin reviews and resolves or rejects the report.

Dashboard stats, content moderation (hide/lock/pin), review author changes, and user management.

End-to-end setup of a complete clinic page including doctors, treatments, reviews, photos, articles, and forum posts.

AI agent creates multiple forum posts on behalf of different users. Requires admin token.

AI agent creates multiple comments and replies on behalf of different users. Requires admin token.

Create a forum post with embedded YouTube/Vimeo video via iframe in HTML content.

Convert existing API docs into visual SOP workflows. Write markdown SOPs, convert to .osop, validate, publish to SOP Doc.

Feed two .osoplog files to osop diff. See per-step duration, cost, and status changes. Know instantly what improved and what broke.

Convert GitHub Actions, Airflow, n8n, CrewAI, Argo, or LangGraph workflows into .osop format. Your existing processes become SOP Doc entries.

Run an API SOP workflow end-to-end. Real HTTP requests with headers, body, auth. Cost controls. Security gates. Execution log generated.

How SOP Doc works. Write .osop files, browse them visually, share with anyone. Non-engineers can read the Visual tab.

Execute .osop → record .osoplog → AI synthesizes better .osop → diff to verify → repeat. Each cycle gets faster, cheaper, more reliable.

S3 → validate → transform → Snowflake → dashboard refresh with failure alerts.

Data prep → train → evaluate → A/B test → deploy to serving endpoint.

Structured debate with pro/con agents, moderator synthesis, and human judgment.

Three-agent crew that researches a topic, analyzes findings, and writes a report.

Question → retrieve → rerank → generate answer with citations → feedback.

Lint → parallel tests → Docker build → staging → approval → production.

Intent-based routing with specialist agents and human escalation.

Webhook-triggered pipeline that enriches, scores, and routes inbound leads.

Fetch PR diff → parallel analysis → generate review → post comments.

Durable order processing with compensation on failure.

Full i18n coverage for the OSOP website: translated all UI components, page content, YAML showcases, and 245 workflow gallery entries from English to Traditional Chinese.