davila7--claude-code-templates
wehub-resource-sync
1 行
5.2 KiB
JSON
1 行
5.2 KiB
JSON
{"content": "---\nname: codex\ndescription: Use when the user asks to run Codex CLI (codex exec, codex resume) or references OpenAI Codex for code analysis, refactoring, or automated editing. Uses GPT-5.2 by default for state-of-the-art software engineering.\n---\n\n# Codex Skill Guide\n\n## Running a Task\n1. Default to `gpt-5.2` model. Ask the user (via `AskUserQuestion`) which reasoning effort to use (`xhigh`,`high`, `medium`, or `low`). User can override model if needed (see Model Options below).\n2. Select the sandbox mode required for the task; default to `--sandbox read-only` unless edits or network access are necessary.\n3. Assemble the command with the appropriate options:\n - `-m, --model <MODEL>`\n - `--config model_reasoning_effort=\"<high|medium|low>\"`\n - `--sandbox <read-only|workspace-write|danger-full-access>`\n - `--full-auto`\n - `-C, --cd <DIR>`\n - `--skip-git-repo-check`\n3. Always use --skip-git-repo-check.\n4. When continuing a previous session, use `codex exec --skip-git-repo-check resume --last` via stdin. When resuming don't use any configuration flags unless explicitly requested by the user e.g. if he species the model or the reasoning effort when requesting to resume a session. Resume syntax: `echo \"your prompt here\" | codex exec --skip-git-repo-check resume --last 2>/dev/null`. All flags have to be inserted between exec and resume.\n5. **IMPORTANT**: By default, append `2>/dev/null` to all `codex exec` commands to suppress thinking tokens (stderr). Only show stderr if the user explicitly requests to see thinking tokens or if debugging is needed.\n6. Run the command, capture stdout/stderr (filtered as appropriate), and summarize the outcome for the user.\n7. **After Codex completes**, inform the user: \"You can resume this Codex session at any time by saying 'codex resume' or asking me to continue with additional analysis or changes.\"\n\n### Quick Reference\n| Use case | Sandbox mode | Key flags |\n| --- | --- | --- |\n| Read-only review or analysis | `read-only` | `--sandbox read-only 2>/dev/null` |\n| Apply local edits | `workspace-write` | `--sandbox workspace-write --full-auto 2>/dev/null` |\n| Permit network or broad access | `danger-full-access` | `--sandbox danger-full-access --full-auto 2>/dev/null` |\n| Resume recent session | Inherited from original | `echo \"prompt\" \\| codex exec --skip-git-repo-check resume --last 2>/dev/null` (no flags allowed) |\n| Run from another directory | Match task needs | `-C <DIR>` plus other flags `2>/dev/null` |\n\n## Model Options\n\n| Model | Best for | Context window | Key features |\n| --- | --- | --- | --- |\n| `gpt-5.2-max` | **Max model**: Ultra-complex reasoning, deep problem analysis | 400K input / 128K output | 76.3% SWE-bench, adaptive reasoning, $1.25/$10.00 |\n| `gpt-5.2` ⭐ | **Flagship model**: Software engineering, agentic coding workflows | 400K input / 128K output | 76.3% SWE-bench, adaptive reasoning, $1.25/$10.00 |\n| `gpt-5.2-mini` | Cost-efficient coding (4x more usage allowance) | 400K input / 128K output | Near SOTA performance, $0.25/$2.00 |\n| `gpt-5.1-thinking` | Ultra-complex reasoning, deep problem analysis | 400K input / 128K output | Adaptive thinking depth, runs 2x slower on hardest tasks |\n\n**GPT-5.2 Advantages**: 76.3% SWE-bench (vs 72.8% GPT-5), 30% faster on average tasks, better tool handling, reduced hallucinations, improved code quality. Knowledge cutoff: September 30, 2024.\n\n**Reasoning Effort Levels**:\n- `xhigh` - Ultra-complex tasks (deep problem analysis, complex reasoning, deep understanding of the problem)\n- `high` - Complex tasks (refactoring, architecture, security analysis, performance optimization)\n- `medium` - Standard tasks (refactoring, code organization, feature additions, bug fixes)\n- `low` - Simple tasks (quick fixes, simple changes, code formatting, documentation)\n\n**Cached Input Discount**: 90% off ($0.125/M tokens) for repeated context, cache lasts up to 24 hours.\n\n## Following Up\n- After every `codex` command, immediately use `AskUserQuestion` to confirm next steps, collect clarifications, or decide whether to resume with `codex exec resume --last`.\n- When resuming, pipe the new prompt via stdin: `echo \"new prompt\" | codex exec resume --last 2>/dev/null`. The resumed session automatically uses the same model, reasoning effort, and sandbox mode from the original session.\n- Restate the chosen model, reasoning effort, and sandbox mode when proposing follow-up actions.\n\n## Error Handling\n- Stop and report failures whenever `codex --version` or a `codex exec` command exits non-zero; request direction before retrying.\n- Before you use high-impact flags (`--full-auto`, `--sandbox danger-full-access`, `--skip-git-repo-check`) ask the user for permission using AskUserQuestion unless it was already given.\n- When output includes warnings or partial results, summarize them and ask how to adjust using `AskUserQuestion`.\n\n## CLI Version\n\nRequires Codex CLI v0.57.0 or later for GPT-5.2 model support. The CLI defaults to `gpt-5.2` on macOS/Linux and `gpt-5.2` on Windows. Check version: `codex --version`\n\nUse `/model` slash command within a Codex session to switch models, or configure default in `~/.codex/config.toml`.\n"} |