{"slug": "theo-s-approach-to-running-gpt-5-5-as-a-claude-code-subagent", "title": "Theo's approach to running GPT-5.5 as a Claude Code subagent", "summary": "Theo Browne published a guide for running GPT-5.5 as a subagent within Claude Code workflows, ranking models by cost, intelligence, and taste. The approach uses GPT-5.5 for bulk mechanical work via the Codex CLI, while reserving Claude models for user-facing tasks requiring higher taste scores. Browne provides detailed configuration instructions for integrating GPT-5.5 into Claude workflows, including wrapper agents and parallel execution with worktree isolation.", "body_md": "(stolen from Theo Browne)\n\nRankings, higher = better. Cost reflects what I actually pay (OpenAI is near-free for me due to a deal), not list price. Intelligence is how hard a problem you can hand the model unsupervised. Taste covers UI/UX, code quality, API design, and copy.\n\n| model | cost | intelligence | taste |\n|---|---|---|---|\n| gpt-5.5 | 9 | 8 | 5 |\n| sonnet-5 | 5 | 5 | 7 |\n| opus-4.8 | 4 | 7 | 8 |\n| fable-5 | 2 | 9 | 9 |\n\nHow to apply:\n\n- These are defaults, not limits. You have standing permission to override them: if a cheaper model's output doesn't meet the bar, rerun or redo the work with a smarter model without asking. Judge the output, not the price tag. Escalating costs less than shipping mediocre work.\n- Cost is a tie-breaker only; when axes conflict for anything that ships, intelligence > taste > cost.\n- Don't let cost prevent you from using the right model for the job. Instead, take advantage of cheaper options to get more information and try things before moving the work to a more expensive option.\n- Bulk/mechanical work (clear-spec implementation, data analysis, migrations): gpt-5.5 — it's effectively free.\n- Anything user-facing (UI, copy, API design) needs taste ≥ 7.\n- Reviews of plans/implementations: fable-5 or opus-4.8, optionally gpt-5.5 as an extra independent perspective.\n- Never use Haiku.\n- Mechanics: gpt-5.5 is only reachable through the Codex CLI —\n`codex exec`\n\n/`codex review`\n\n(my ~/.codex/config.toml defaults to gpt-5.5). Use the codex-implementation, codex-review, and codex-computer-use skills; for work they don't cover (investigation, data analysis), run`codex exec -s read-only`\n\ndirectly with a self-contained prompt. - Claude models (sonnet-5, opus-4.8, fable-5) run via the Agent/Workflow model parameter. Using gpt-5.5 inside workflows and subagents (the model parameter only takes Claude models, so use a wrapper):\n- Spawn a thin Claude wrapper agent with\n`model: 'sonnet', effort: 'low'`\n\nwhose prompt instructs it to write a self-contained codex prompt, run`codex exec`\n\nvia Bash, and return the report (use`schema`\n\non the wrapper to get structured output back). - Always label these agents with a\n`gpt-5.5:`\n\nprefix, e.g.`{label: 'gpt-5.5:review-auth'}`\n\n— the workflow UI shows the wrapper's Claude model, so the label is the only indication the real worker is gpt-5.5. - Codex runs can exceed Bash's 10-minute timeout: pass an explicit timeout, or run in the background and poll for the report file.\n- Parallel gpt-5.5 implementation agents must use\n`isolation: 'worktree'`\n\nso codex edits don't collide in the shared checkout. - Workflow token budgets only count Claude tokens; codex work is free and invisible to\n`budget.spent()`\n\n.", "url": "https://wpnews.pro/news/theo-s-approach-to-running-gpt-5-5-as-a-claude-code-subagent", "canonical_source": "https://gist.github.com/jacobparis/94d5cd3a60bc55b2a1afa3a428d0523a", "published_at": "2026-07-07 03:58:06+00:00", "updated_at": "2026-07-11 17:39:33.121106+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-tools", "ai-agents", "developer-tools"], "entities": ["Theo Browne", "GPT-5.5", "Claude Code", "Codex CLI", "sonnet-5", "opus-4.8", "fable-5"], "alternates": {"html": "https://wpnews.pro/news/theo-s-approach-to-running-gpt-5-5-as-a-claude-code-subagent", "markdown": "https://wpnews.pro/news/theo-s-approach-to-running-gpt-5-5-as-a-claude-code-subagent.md", "text": "https://wpnews.pro/news/theo-s-approach-to-running-gpt-5-5-as-a-claude-code-subagent.txt", "jsonld": "https://wpnews.pro/news/theo-s-approach-to-running-gpt-5-5-as-a-claude-code-subagent.jsonld"}}