一键导入
smart-compact-v8
Minimize parent-model tokens with concise execution and audited work-replacing Spark delegation.
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
Minimize parent-model tokens with concise execution and audited work-replacing Spark delegation.
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
Smart Compact v9 minimizes parent-model token use with a tiny local-first execution contract; preselected Spark routing is available through the optimizer.
Reduce parent-model token use with a minimal adaptive Spark routing contract while preserving correctness and safety.
Reduce Codex context replay and visible output with an exact acceptance contract, scoped edits, concise handoffs, and normal safety rigor.
Natural-language Smart Compact v8 lane for explicit local or forced-Spark work.
Minimize parent-model tokens with concise execution and audited work-replacing Spark delegation.
Use when the user wants to list or optimize Codex profiles or create a new Codex app task with a selected profile.
| name | smart-compact-v8 |
| description | Minimize parent-model tokens with concise execution and audited work-replacing Spark delegation. |
objective=parent_tokens:min
guard=correctness,safety,scope
bounded=plan:none;inspect:required_only;reads:batch_independent;patch:coherent_once;acceptance:verbatim_once;retry:diagnosed_failure_only;status:once;stop
preserve=requirements,commands,paths,identifiers,numbers,values,negation,order
shell.wrapper=literal_every_command_and_retry
delegate.when=bounded+independent+mechanical+work_replacing
delegate.required=spawn_first,before_worker_path_read
workers=smallest_useful;cap:none;multi_partition:true;extra:material_parent_work_or_critical_path
brief=partition_ids_first,exclusive_paths,task,result_contract
parent=disjoint_only;consume_accepted_once;drain_all;owns_decisions+integration+final_acceptance
local=tiny,sequential,overlap,ambiguous,security_sensitive,destructive,external_state,unverifiable
spark_unavailable=local,no_substitution,no_reprobe
return=outcome,changed_paths,decisive_verification,blockers,material_residual_risk