{"id":31738131,"url":"https://github.com/panaversity/learn-agentic-ai-from-low-code-to-code","last_synced_at":"2025-10-14T14:01:55.755Z","repository":{"id":318609873,"uuid":"1072000538","full_name":"panaversity/learn-agentic-ai-from-low-code-to-code","owner":"panaversity","description":"Build production-grade agents with OpenAI AgentKit, the Apps SDK, Agents SDK, MCP, LLM CLIs, and Spec-Kit Plus.","archived":false,"fork":false,"pushed_at":"2025-10-08T08:15:40.000Z","size":7649,"stargazers_count":3,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-10-08T08:16:19.601Z","etag":null,"topics":["agent-kit","agents-sdk","apps-sdk","mcp","openai"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/panaversity.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-10-08T05:55:01.000Z","updated_at":"2025-10-08T08:15:44.000Z","dependencies_parsed_at":"2025-10-08T08:16:22.700Z","dependency_job_id":"6ef0c9b9-36d4-4bff-a19f-43f7ded3bb44","html_url":"https://github.com/panaversity/learn-agentic-ai-from-low-code-to-code","commit_stats":null,"previous_names":["panaversity/learn-agentic-ai-from-low-code-to-code"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/panaversity/learn-agentic-ai-from-low-code-to-code","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/panaversity%2Flearn-agentic-ai-from-low-code-to-code","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/panaversity%2Flearn-agentic-ai-from-low-code-to-code/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/panaversity%2Flearn-agentic-ai-from-low-code-to-code/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/panaversity%2Flearn-agentic-ai-from-low-code-to-code/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/panaversity","download_url":"https://codeload.github.com/panaversity/learn-agentic-ai-from-low-code-to-code/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/panaversity%2Flearn-agentic-ai-from-low-code-to-code/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":279019051,"owners_count":26086518,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-10-14T02:00:06.444Z","response_time":60,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agent-kit","agents-sdk","apps-sdk","mcp","openai"],"created_at":"2025-10-09T09:52:45.424Z","updated_at":"2025-10-14T14:01:55.583Z","avatar_url":"https://github.com/panaversity.png","language":null,"funding_links":[],"categories":["📚 Projects (1974 total)"],"sub_categories":["MCP Servers"],"readme":"# Learn Agentic AI: From Low-Code to Code\n*Build production-grade agents with OpenAI AgentKit*\n\nThis Free Course is Offered by: [Panaversity](https://panaversity.org/)\n\n**Register in this Free Panaversity Course:** [Click to Register](https://panaversity.org/flagship-program/courses/AI-100)\n\n**All-Pakistan and All-Star Faculty:**\nQasim (Karachi), Ameen (Saudi Arabia), Wania (Wah Cantt.), Junaid (Lahore), Aneeq (Karachi), Hammad (Peshawar), Rehan (Islamabad) and Zia (Karachi).\n\n**Live YouTube Classes [@panaversity channel](https://www.youtube.com/@panaversity) on Wednesday and Thursday and 8:00 pm - 10:00 pm:** Class starting from Oct 15, 2025. Recording Also Available on [@Panaversity YouTube Channel](https://www.youtube.com/@panaversity)\n\nHere’s a practical, non-coding syllabus which we are running as a 4-week cohort (8 x 2–3h sessions) or we will also offer it as a 2-day intensive in Companies and Universities. It stays hands-on with AgentKit’s visual tools—no programming required.\n\n## No-Code Agents with OpenAI AgentKit — Course Syllabus\n\n## Course overview\n\nBuild, test, and ship a production-ready AI agent using OpenAI’s new AgentKit—focusing on the visual Agent Builder, governance \u0026 guardrails, built-in evals, and one-click chat UI embedding with ChatKit. By the end, learners publish a working agent and a simple plan to measure impact and iterate. ([OpenAI][1])\n\n### Who it’s for\n\nNon-programmers (general managers, accountants, engineers, doctors, financial managers, product managers, ops, Customer Experience (CX), Human Resources / Learning \u0026 Development (HR/L\u0026D), consultants, educators, everyone) who can define workflows and business outcomes, but don’t want to write code.\n\n### What you’ll build\n\nA capstone agent of your choice—for example:\n\n* Customer support triage \u0026 answer bot\n* Research \u0026 brief generator for sales/BD\n* Internal knowledge assistant (policies/SOPs)\n* Classroom co-teacher / onboarding guide\n\n---\n\n## Learning outcomes\n\nBy the end, participants can:\n\n1. Understand the concepts of Agentic Web and Agentic Organizations\n2. Describe core agent concepts (agents, tools, handoffs, sessions, guardrails) in plain language and map them to business workflows. ([OpenAI GitHub][2])\n3. Use **Agent Builder** to design multi-step workflows on a visual canvas, version them, and add approvals/guardrails without code. ([OpenAI][1])\n4. Connect knowledge safely (files/connectors) and set up the **Connector Registry** with admins. ([OpenAI][1])\n5. Deploy a branded chat experience with **ChatKit** and share it with test users. ([OpenAI][3])\n6. Measure and improve quality with **Evals** (datasets, trace grading, prompt optimization), and understand when to consider reinforcement fine-tuning. ([OpenAI][1])\n7. Apply **Guardrails** for safety (PII masking, jailbreak detection, hallucination checks) using the wizard and presets. ([OpenAI GitHub][4])\n\n---\n\n## Format \u0026 pacing\n\n* **6 weeks** · **2 sessions/week** · **2–3 hours/session** (+ extra Question/Answer Sessions)\n* **Alt**: **2-day bootcamp** covering the same modules with condensed labs to be offered in Companies and Universities\n* Delivery: live workshop or flipped classroom (short videos + guided labs)\n\n---\n\n## Detailed Schedule\n\n### Week 1 — Agentic Web and Organizations \n\n**Session 1: [Agentic Web](./00_agentic_web/readme.md) (2h)**\n\n**Session 2: [Agentic Organizations](./01_agentic_org/readme.md) (2h)**\n\n### Week 2 — Foundations \u0026 first workflow\n\n**Session 1: What is an “agent” (no code)? (2h)**\n\n* The AgentKit stack at a glance: Agent Builder (visual), Connector Registry (admin), ChatKit (UI), Evals (quality), Guardrails (safety).\n* Concepts in human terms: tasks, tools, multi-step flows, agent handoffs, memory/sessions.\n* Demo tour of Agent Builder: canvases, nodes, versions, templates; publish/preview lifecycle.\n  \n  **Lab:** Clone a template, customize instructions, add an approval step, run test conversations. ([OpenAI][1])\n\n**Session 2: Connecting knowledge safely (2–3h)**\n\n* What to put in vs. link to; file search basics; connector options (e.g., Drive, SharePoint, Teams) and MCP servers.\n* Admin view of **Connector Registry**; roles \u0026 governance; enabling connectors for a workspace.\n* Data handling patterns: least-privilege, redaction, auditability.\n  \n  **Lab:** Attach a small policy pack (PDFs/Docs), configure retrieval, and test relevance safely. ([OpenAI][1])\n\n### Week 3 — Designing, deploying, and branding the chat UI\n\n**Session 3: Visual design patterns (2–3h)**\n\n* Drag-and-drop nodes: tools, file search, guardrails, decision/branching, human-in-the-loop.\n* Multi-agent patterns via handoffs—when and why to split responsibilities.\n* Versioning \u0026 change logs; rollback and safe launches.\n  \n  **Lab:** Build a 5–7 node workflow from a blank canvas; add a human approval and a fallback path. ([OpenAI][1])\n\n**Session 4: Deploy with ChatKit (2h)**\n\n* Shipping a usable interface without front-end work: embed options and theme/brand tweaks.\n* Sharing with pilot users; capturing transcripts and feedback for iteration.\n  \n  **Lab:** Deploy your agent’s chat UI, set a custom name/avatar, and invite 3 pilot testers. ([OpenAI][3])\n\n### Week 4 — Quality: evaluate, observe, and iterate\n\n**Session 5: Evals you’ll actually use (2–3h)**\n\n* Designing a simple eval dataset from real tickets/prompts.\n* **Trace grading**: grading whole runs to spot brittle steps.\n* **Prompt optimization**: generate improved prompts from grader + human annotations.\n* Third-party model comparisons (what, why, when).\n  \n  **Lab:** Create a 20-case eval, run it, and apply one optimization round. Re-run and compare. ([OpenAI][1])\n\n**Session 6: Observability \u0026 cost control (2h)**\n\n* Reading traces; identifying tool-call loops and dead ends.\n* Lightweight A/Bs: instruction tweaks and guardrail thresholds.\n* When RFT helps (conceptual only) and how to scope an RFT request with your tech team.\n  \n  **Lab:** Use traces to remove one redundant step and reduce tokens/time per task. ([OpenAI][1])\n\n### Week 5 — Safety, governance, and capstone\n\n**Session 7: Guardrails \u0026 governance (2–3h)**\n\n* The **Guardrails Wizard**: select checks (moderation, jailbreak, PII, hallucinations) and set policies—no code.\n* Human approvals \u0026 audit trails; rollout controls and change management.\n  \n  **Lab:** Add PII masking + jailbreak detection to your agent, document your policy, and re-test. ([OpenAI GitHub][4])\n\n**Session 8: Capstone build \u0026 publish (2–3h)**\n\n* Finalize workflow, run evals, harden guardrails, and polish the ChatKit UI.\n* Write a one-page “launch note”: scope, metrics, SLAs, rollback plan.\n  \n  **Capstone demo:** 5-minute live run + Q\u0026A; submit your Launch Note and a 14-day iteration plan.\n\n---\n\n## Assessments \u0026 deliverables\n\n* **Checkpoints (30%)**: Session labs (Builder, connectors, evals, guardrails).\n* **Capstone (50%)**: Working agent, embedded chat UI, eval results, safety config.\n* **Launch Note (20%)**: Clear goals, success metrics, and iteration plan.\n\n---\n\n## Tools \u0026 accounts (no coding needed)\n\n* **Agent Builder** (visual canvas, versioning, preview/publish)\n* **Connector Registry** (admin-managed app \u0026 data connections across orgs)\n* **ChatKit** (embedded chat UI with branding controls)\n* **Evals** (datasets, trace grading, prompt optimization, optional third-party models)\n* **Guardrails** (wizard + presets for PII/jailbreak/hallucination checks)\n  Availability: as of **Oct 6–8, 2025**, Agent Builder is in beta; ChatKit and new Evals features are generally available; Connector Registry is rolling out in beta via the Global Admin Console. All are included under standard API model pricing. ([OpenAI][1])\n\n---\n\n## Instructor prep \u0026 room setup\n\n* Learners need: laptop, browser, sample documents (FAQs, policies, SOPs), and access granted to AgentKit features in your org.\n* Create an org sandbox + test connectors; prepare three Builder templates per track (support, research, internal knowledge).\n\n---\n\n## Optional 2-day bootcamp agenda (condensed) to be offered in Universities and Companies\n\n**Day 1 AM:** Foundations + first workflow (Sessions 1–3)\n**Day 1 PM:** Data connections + ChatKit deploy (Session 4)\n**Day 2 AM:** Evals + optimization (Sessions 5–6)\n**Day 2 PM:** Guardrails + capstone launch (Sessions 7–8)\n\n\n---\n\n## Four learning levels\n\nAfter completing this course (Level 1) you will be at Learning Level 1, after that you will be ready for Levels 2, 3, and 4.\n\n* **Learning Level 1: No-Code Agent Development** — build end-to-end in AgentKit (This Course)\n* **Learning Level 2: Code-First Chat GPT Apps SDK** — [Learn Python](https://github.com/panaversity/learn-modern-ai-python/tree/main/00_python_colab) and start in Apps SDK, then import visual flows where helpful (This course will be offered by Panaversity later). \n* **Learning Level 3: Full-Code Agents SDK** - [Learn Agentic AI using OpenAI Agents SDK and MCP](https://github.com/panaversity/learn-agentic-ai)\n* **Learning Level 4: Full-Code AI Assisted: Spec-Driven Vibe-Coding** - [Spec-Kit Plus](https://github.com/panaversity/spec-kit-plus)\n\n\n---\n\n\n### References\n\n* OpenAI **Introducing AgentKit** (features, availability, pricing). ([OpenAI][1])\n* OpenAI **Agent platform overview** (Builder, ChatKit, Evals). ([OpenAI][3])\n* **Agents SDK** docs (background on agents/handoffs/guardrails—used conceptually here). ([OpenAI GitHub][2])\n* **Guardrails** docs (wizard \u0026 presets). ([OpenAI GitHub][4])\n* Launch coverage \u0026 context. ([TechCrunch][5])\n\nIf you want, I can tailor this to a specific audience (e.g., support orgs, schools, or consulting firms) and pre-fill the Builder templates for that track.\n\n[1]: https://openai.com/index/introducing-agentkit/ \"Introducing AgentKit | OpenAI\"\n[2]: https://openai.github.io/openai-agents-python/ \"OpenAI Agents SDK\"\n[3]: https://openai.com/agent-platform/ \"API Agents | OpenAI\"\n[4]: https://openai.github.io/openai-guardrails-python/ \"OpenAI Guardrails Python\"\n[5]: https://techcrunch.com/2025/10/06/openai-launches-agentkit-to-help-developers-build-and-ship-ai-agents/?utm_source=chatgpt.com \"OpenAI launches AgentKit to help developers build and ship AI agents\"\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpanaversity%2Flearn-agentic-ai-from-low-code-to-code","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpanaversity%2Flearn-agentic-ai-from-low-code-to-code","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpanaversity%2Flearn-agentic-ai-from-low-code-to-code/lists"}