SyntaxPilot
Courses/OpenClaw: Build Your Own AI Agent

Advanced · Free course

OpenClaw: Build Your Own AI Agent

How an AI agent harness works, file by file, and how to build your own.

60 lessons0 videos so farNotes for every lesson

What you’ll learn

  • The agent loop
  • Tools and memory
  • The OpenClaw harness
  • Your personal agent
0 of 60 complete

Curriculum

  1. 1What an AI agent is, and what OpenClaw issoon
  2. 2The big picture: OpenClaw's architecturesoon
  3. 3Installing OpenClawsoon
  4. 4Model providers and API keys, safelysoon
  5. 5Onboarding and your first conversationsoon
  6. 6The agent workspacesoon
  7. 7Configurationsoon
  8. 8Models, thinking and failover basicssoon
  9. 9What a harness issoon
  10. 10The agent loop I: one turnsoon
  11. 11The agent loop II: tool callssoon
  12. 12Streamingsoon
  13. 13Tool batches and parallel toolssoon
  14. 14When the model gets tools wrong: recovery and repairsoon
  15. 15Guards against runaway agentssoon
  16. 16Interruption and steeringsoon
  17. 17The system prompt, assembledsoon
  18. 18Context and the context enginesoon
  19. 19Compaction I: when and where to cutsoon
  20. 20Compaction II: summarisingsoon
  21. 21Sessions and transcriptssoon
  22. 22The session controller: one writer at a timesoon
  23. 23A run from start to finishsoon
  24. 24Errors, retries and model failoversoon
  25. 25Shaping the replysoon
  26. 26Hooks I: internal hookssoon
  27. 27Hooks II: plugin hookssoon
  28. 28Reasoning and thinkingsoon
  29. 29Agent runtimessoon
  30. 30Plugging in your own harnesssoon
  31. 31Debugging a run: trajectories and logssoon
  32. 32Usage and cost trackingsoon
  33. 33Build a mini harness I: the loop and one toolsoon
  34. 34Build a mini harness II: streaming, parallel tools, interruptionsoon
  35. 35Build a mini harness III: transcript and compactionsoon
  36. 36Built-in toolssoon
  37. 37exec and exec approvalssoon
  38. 38Sandboxingsoon
  39. 39Skillssoon
  40. 40Write your first skillsoon
  41. 41MCP in OpenClawsoon
  42. 42Pluginssoon
  43. 43Memory Isoon
  44. 44Memory IIsoon
  45. 45Sub-agents and multi-agent routingsoon
  46. 46Secrets and the security modelsoon
  47. 47An agent that acts on its ownsoon
  48. 48The gatewaysoon
  49. 49Channelssoon
  50. 50Clients, Control UI and the HTTP APIssoon
  51. 51Nodes and devicessoon
  52. 52Running it somewhere elsesoon
  53. 53Watching it in productionsoon
  54. 54Your personal agent I: the designsoon
  55. 55Your personal agent II: workspace, model and memorysoon
  56. 56Your personal agent III: skills for daily worksoon
  57. 57Your personal agent IV: a custom tool plugin and hookssoon
  58. 58Your personal agent V: Telegram, approvals and sandboxsoon
  59. 59Your personal agent VI: testing and evaluationsoon
  60. 60Your personal agent VII: running it 24/7, safelysoon