一键导入
configure-eve-driver
Use when choosing, editing, or overriding EvE driver settings for codex_smoke, codex_max, or interactive/debug runs.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Use when choosing, editing, or overriding EvE driver settings for codex_smoke, codex_max, or interactive/debug runs.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when creating new skills, editing existing skills, or verifying skills work before deployment
Use when creating new subagents, editing existing subagents, or verifying subagents work before deployment
Use when an Eve agent is behaving wrong (stuck, gaming the score, bad reasoning) and you need to read its own rollout to find out why
Use when implementing a sanity check subagent for the phase 2 optimization in Eve loop.
Use when implementing a subagent.
Use when creating new skills, editing existing skills, or verifying skills work before deployment
| name | configure-eve-driver |
| description | Use when choosing, editing, or overriding EvE driver settings for codex_smoke, codex_max, or interactive/debug runs. |
Use the public presets first:
driver: codex_smoke is for fast runtime validation.driver: codex_max is for full non-interactive runs.Do not add a new preset for a one-off variation. Prefer Hydra overrides until the role is stable enough to document as a supported entrypoint.
Driver presets live in configs/eve/driver/. Keep that directory small:
codex_smoke.yaml: non-interactive Codex backend, small turn budget, web search disabled.codex_max.yaml: non-interactive Codex backend, full-run budget, web search enabled.Task smoke configs live beside task configs, but keep the root config surface compact:
circle_packing.smoke is the single tracked circle packing smoke preset.Task smoke configs should use driver: codex_smoke. Do not shrink loop.n_workers_phase2 in the default smoke path; the two-worker default covers Phase 2 multi-worker behavior and optimizer sync.
Fast smoke:
uv run python -m scaling_evolve.algorithms.eve.runner --config-name=circle_packing.smoke
Judge evaluation smoke:
uv run python -m scaling_evolve.algorithms.eve.runner \
--config-name=circle_packing.smoke \
evaluation=circle_packing.judge \
label=circle-packing-judge-smoke
Role-specific system prompt smoke:
uv run python -m scaling_evolve.algorithms.eve.runner \
--config-name=circle_packing.smoke \
evaluation=circle_packing.judge \
loop.max_iterations=1 \
label=circle-packing-codex-prompt-smoke \
+driver.overrides.solver.system_prompt_file=configs/eve/optimizer/circle_packing/prompt/CODEX_SYSTEM_PROMPT.md
Full run:
uv run python -m scaling_evolve.algorithms.eve.runner --config-name=circle_packing
One-off interactive/debug run:
uv run python -m scaling_evolve.algorithms.eve.runner \
--config-name=circle_packing.smoke \
driver.driver=codex_tmux \
driver.open_iterm2=true
If overriding from a full config, keep the backend switch explicit:
uv run python -m scaling_evolve.algorithms.eve.runner \
--config-name=circle_packing \
driver.driver=codex_tmux \
driver.open_iterm2=true
Do not preserve interactive/debug overrides as new YAML presets unless the team decides they are a supported public entry.
The driver config lives under the top-level driver: key.
driver.driver selects the backend and wins over driver.provider.driver.provider is accepted for older configs. Avoid it for new public presets.codex_max, changing only driver.provider will not switch the backend because driver.driver=codex_exec is still set.driver mapping, then overlays driver.overrides.solver or driver.overrides.eval.driver.overrides.optimizer; the independent optimizer driver role was removed.Supported backend names are codex_exec and codex_tmux. The public presets should still remain codex_smoke and codex_max.
Common Codex-style fields (codex_exec, codex_tmux):
driver.executable: command name to launch.driver.model: model identifier passed to the backend.driver.reasoning_effort: reasoning budget where supported.driver.effort_level: accepted as a fallback spelling for Codex-style reasoning effort.driver.rollout_max_turns: per-rollout agent turn cap.driver.timeout_seconds: per-rollout wall-clock timeout.driver.web_search: disabled, cached, or live.driver.budget_prompt: whether EvE injects the budget prompt wrapper.driver.enable_multi_agent: optional backend feature override.driver.personality: optional backend personality override.driver.system_prompt_file: repo-relative path to a backend system prompt file.driver.model_provider: optional alternate provider id for Codex-style backends.driver.model_providers: provider config map; if an entry declares env_key, EvE fails fast when that environment variable is missing.driver.token_pricing: per-driver pricing metadata override. Prefer shared pricing in configs/pricing.yaml unless a run needs a local override.Tmux-only fields (codex_tmux):
driver.approval_policy: approval policy passed to the interactive backend.driver.sandbox_mode: sandbox mode passed to the interactive backend.driver.allow_network: when sandbox_mode is omitted, true maps to danger-full-access; otherwise the default is workspace-write.driver.pool_size: shared tmux pane pool size when any role uses a tmux backend. Default is the worker count.driver.open_iterm2: whether EvE opens the tmux session in a visible terminal window.driver.completion_filename: internal tmux completion marker filename.driver.instruction_filename: internal tmux instruction filename.This skill is grounded in the current runtime surfaces:
src/scaling_evolve/algorithms/eve/runtime/driver.py for backend selection, role overrides, tmux pool creation, sandbox/web-search normalization, provider environment resolution, and pricing config loading.src/scaling_evolve/providers/agent/drivers/codex_exec.py for non-interactive Codex fields.src/scaling_evolve/providers/agent/drivers/codex_tmux.py for interactive Codex/tmux fields.src/scaling_evolve/providers/agent/config.py for persistent session provider config validation.