菜单
Report frugal routing metrics - cost per agent tier, escalation rate, and estimated savings versus running everything on the top-tier model.
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
Cost-optimised task routing. Use at the start of any coding, search, extraction, review, or multi-step task to pick the cheapest execution strategy (deterministic tool, haiku worker, sonnet worker, main model, fable worker) and to handle escalation when a worker fails. Also triggers on mentions of cost, budget, routing, or delegation.
Show or change which model each frugal agent runs on, per project. Use when the user wants to see the current tier-to-model mapping, move an agent to a different model (e.g. scout to sonnet), or reset overrides. Triggers on "frugal models", "change router models", "which model does scout use".
Wire the frugal savings segment into the user's Claude Code statusline - creates a minimal statusline if none exists, or merges the segment into an existing one.
| name | router-stats |
| description | Report frugal routing metrics - cost per agent tier, escalation rate, and estimated savings versus running everything on the top-tier model. |
| disable-model-invocation | true |
Run the report and show its output verbatim as markdown:
python3 "<plugin-root>/scripts/stats.py"
<plugin-root> is two directories up from this skill's base directory (the base directory is announced when this skill loads). Pass --path <file> only if the user names a different metrics file.
After the table: if any agent's escalation count exceeds 20% of its runs, point at the routing decision table row that routes to it and suggest moving that task type one tier up.