{
  "schemaVersion": "1",
  "revision": "22ce5d34dbcb5a78",
  "id": "codex-workflow-selection",
  "slug": "workflow-selection",
  "title": "Choose the next agent workflow",
  "summary": "Provides a bottleneck-driven agent prompt and an eleven-step workflow connecting context, shell visibility, routing, composition, verified variations, evidence, delegation, council critique, and human selection.",
  "category": "Planning",
  "tags": [
    "workflow",
    "bottlenecks",
    "decisions"
  ],
  "sources": [
    {
      "label": "Portable Codex workflow resources",
      "url": "https://lapis-fossil-vsbc.here.now/downloads/codex-power-user-workshop-2026-06-27.bundle"
    }
  ],
  "sourceUnits": [
    "downloads/codex-power-user-workshop-2026-06-27.bundle!/cheatsheets/00-index.md",
    "downloads/codex-power-user-workshop-2026-06-27.bundle!/cheatsheets/00-index.md#00. Workshop Workflow Index",
    "downloads/codex-power-user-workshop-2026-06-27.bundle!/cheatsheets/00-index.md#00. Workshop Workflow Index/Agent Setup Prompt",
    "downloads/codex-power-user-workshop-2026-06-27.bundle!/cheatsheets/00-index.md#00. Workshop Workflow Index/Phase Map",
    "downloads/codex-power-user-workshop-2026-06-27.bundle!/cheatsheets/00-index.md#00. Workshop Workflow Index/Phase Map/Phase 1: Context And Environment",
    "downloads/codex-power-user-workshop-2026-06-27.bundle!/cheatsheets/00-index.md#00. Workshop Workflow Index/Phase Map/Phase 2: Routing And Composition",
    "downloads/codex-power-user-workshop-2026-06-27.bundle!/cheatsheets/00-index.md#00. Workshop Workflow Index/Phase Map/Phase 3: Generation And Evidence",
    "downloads/codex-power-user-workshop-2026-06-27.bundle!/cheatsheets/00-index.md#00. Workshop Workflow Index/Phase Map/Phase 4: Delegation And Review",
    "downloads/codex-power-user-workshop-2026-06-27.bundle!/cheatsheets/00-index.md#00. Workshop Workflow Index/Section Map",
    "downloads/codex-power-user-workshop-2026-06-27.bundle!/cheatsheets/00-index.md#00. Workshop Workflow Index/How To Use The Set",
    "downloads/codex-power-user-workshop-2026-06-27.bundle"
  ],
  "collection": "codex",
  "categoryId": "planning",
  "wordCount": 804,
  "url": "/topics/codex-workflow-selection/",
  "markdownUrl": "/markdown/codex-workflow-selection.md?raw=1&v=22ce5d34dbcb5a78",
  "jsonUrl": "/api/v1/documents/codex-workflow-selection.json",
  "headings": [
    {
      "id": "agent-setup-prompt",
      "text": "Agent Setup Prompt",
      "level": 2
    },
    {
      "id": "phase-map",
      "text": "Phase Map",
      "level": 2
    },
    {
      "id": "phase-1-context-and-environment",
      "text": "Phase 1: Context And Environment",
      "level": 3
    },
    {
      "id": "phase-2-routing-and-composition",
      "text": "Phase 2: Routing And Composition",
      "level": 3
    },
    {
      "id": "phase-3-generation-and-evidence",
      "text": "Phase 3: Generation And Evidence",
      "level": 3
    },
    {
      "id": "phase-4-delegation-and-review",
      "text": "Phase 4: Delegation And Review",
      "level": 3
    },
    {
      "id": "section-map",
      "text": "Section Map",
      "level": 2
    },
    {
      "id": "how-to-use-the-set",
      "text": "How To Use The Set",
      "level": 2
    },
    {
      "id": "related-resources",
      "text": "Related resources",
      "level": 2
    }
  ],
  "markdown": "# Choose the next agent workflow\n\nUse this as the map for the full workflow. Each section stands alone, but the sequence is deliberate.\n\n## Agent Setup Prompt\n\n```text\nUse this workflow index to choose the next agent move for a software project.\n\nFirst identify the current bottleneck:\n- missing context\n- missing shell tools\n- unclear helper behavior\n- wrong workspace or session\n- weak prompt shape\n- too many plausible implementations\n- video or runtime evidence to analyze\n- bounded debug packet to preserve\n- narrow task to delegate\n- risky decision that needs independent model critique\n- generated options that need a human pick\n\nThen pick the matching section and ask the agent to follow that workflow. Keep the request concrete: name the repo, the bug or decision, the files or logs to inspect, and the proof you expect before the work is considered done.\n```\n\n## Phase Map\n\n### Phase 1: Context And Environment\n\nMake invisible project context and agent shell context explicit before asking for work.\n\n1. Pack context when broad code reasoning needs Oracle.\n2. Make the shell and tools visible to Codex.\n3. Classify zsh helper names and fallback boundaries.\n\nTransition: Once the agent can see the right context and tool surface, route the work to the right place before launch.\n\n### Phase 2: Routing And Composition\n\nPut work in the right workspace, pane, session, and prompt shape before it runs.\n\n4. Route the right pane/session.\n5. Compose before launch.\n\nTransition: Once launch is deliberate, use the agent surface to produce, compare, and preserve evidence.\n\n### Phase 3: Generation And Evidence\n\nGenerate useful options and turn rich runtime or media evidence into prompt material.\n\n6. Generate verified variations.\n7. Use video as prompt material.\n8. Mark debug moments with bounded evidence.\n\nTransition: Once the work is bounded and evidenced, delegate narrow packets or ask for independent review.\n\n### Phase 4: Delegation And Review\n\nHand small tool-shaped work to narrow agents, use councils for expensive decisions, and capture human choices before implementation.\n\n9. Hand tool-shaped packets to narrow Imps.\n10. Ask a model council when independent disagreement improves a risky decision.\n11. Capture a phone-friendly human selection from generated options and resume from saved JSON.\n\n## Section Map\n\n| Phase | Section | Workflow Move | File |\n| --- | --- | --- | --- |\n| Context And Environment | 1 | Pack context when broad code reasoning needs Oracle | [01-oracle-packing-context.md](/topics/codex-context-packing-reference/) |\n| Context And Environment | 2 | Make the shell and tools visible to Codex | [02-loading-zsh-env-into-codex.md](/topics/codex-shell-environment-reference/) |\n| Context And Environment | 3 | Classify zsh helper names and fallback boundaries | [03-zsh-functions-aliases.md](/topics/codex-shell-functions-reference/) |\n| Routing And Composition | 4 | Route the right pane/session | [04-agentic-dev-environment-orchestrator.md](/topics/codex-workspace-routing-reference/) |\n| Routing And Composition | 5 | Compose before launch | [05-prompt-builder.md](/topics/codex-prompt-composition-reference/) |\n| Generation And Evidence | 6 | Generate verified variations | [06-code-is-free-generate-variations.md](/topics/codex-variation-workflows-reference/) |\n| Generation And Evidence | 7 | Use video as prompt material | [07-video-as-prompts.md](/topics/codex-video-workflows-reference/) |\n| Generation And Evidence | 8 | Mark debug moments with bounded evidence | [08-debugging-with-agent-markers.md](/topics/codex-marker-workflows-reference/) |\n| Delegation And Review | 9 | Hand tool-shaped packets to narrow Imps | [09-imps.md](/topics/codex-imp-workflows-reference/) |\n| Delegation And Review | 10 | Ask a model council when independent disagreement improves a risky decision | [10-fusion-model-council.md](/topics/codex-fusion-workflows-reference/) |\n| Delegation And Review | 11 | Capture a phone-friendly human selection from generated options | [11-web-choices.md](/topics/codex-choice-workflows-reference/) |\n\n## How To Use The Set\n\nStart with the section that matches the current bottleneck:\n\n| Bottleneck | Start With |\n| --- | --- |\n| The agent needs a broad codebase answer | Section 1 |\n| A command works locally but not in Codex | Section 2 |\n| A helper name is unclear or shell-only | Section 3 |\n| Work is landing in the wrong pane or session | Section 4 |\n| The prompt needs shaping before launch | Section 5 |\n| Several implementations are plausible | Section 6 |\n| A recording contains the missing context | Section 7 |\n| Logs exist but the important moment is unclear | Section 8 |\n| The task is mostly one command-line tool | Section 9 |\n| Several strong models may disagree on a risky architecture or implementation choice | Section 10 |\n| The agent generated good options but needs a human pick or feedback before continuing | Section 11 |\n\nSection 10 is especially useful after Section 6 for comparing strategies, or after Section 9 when a bounded packet needs independent review before a risky implementation choice.\nSection 11 is especially useful after Section 6 or Section 10 when generated options need a human selection, ranking, or feedback before implementation continues.\n\n## Related resources\n\n- [Companion reference](https://lapis-fossil-vsbc.here.now/downloads/codex-power-user-workshop-2026-06-27.bundle)\n- [Resource directory and portable download](/topics/codex-resources/)"
}
