一键导入
smithers-eval
Run a workflow over a JSON/JSONL eval suite and write a regression report. Run `bunx smithers-orchestrator eval --help` for usage details.
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
Run a workflow over a JSON/JSONL eval suite and write a regression report. Run `bunx smithers-orchestrator eval --help` for usage details.
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
Create structured plans for multi-step tasks -- software features, research workflows, events, study plans, or any goal that benefits from breakdown. Also deepens existing plans with interactive sub-agent review. Use when the user says 'plan this', 'create a plan', 'how should we build', 'break this down', or when a brainstorm doc is ready for planning. Use 'deepen the plan' or 'deepening pass' for the deepening flow. For exploratory requests, prefer ce-brainstorm first.
Select the right Bead to work on using Beads robot-mode CLI evidence. Use when asked what Bead to pick next, which local issue is highest leverage, which blocker to clear, or how to choose work from a Beads graph. Produces a validator-checked .laguna/bead-selection.json artifact grounded in bv/br robot output, never the interactive bv TUI.
Inventory the top level of a workspace and write .laguna/workspace-inventory.json. Use when an agent or user needs a quick orientation snapshot: what top-level entries exist, which are files vs directories, and how many files each directory contains. Triggers on requests like "what's at the top of this workspace", "give me a lay of the land", "list the top-level structure", "how many files are in each directory", or any task whose first step is understanding what is present without doing deep analysis. Every claim (entry names, kinds, file counts, total) is mechanically checkable against the real filesystem tree.
Inspect and register subscriptions and api keys. Run `bunx smithers-orchestrator agents --help` for usage details.
List and manage durable alert instances. Run `bunx smithers-orchestrator alerts --help` for usage details.
Approve a paused approval gate. Auto-detects the pending node if only one exists. Run `bunx smithers-orchestrator approve --help` for usage details.
| name | smithers-eval |
| description | Run a workflow over a JSON/JSONL eval suite and write a regression report. Run `bunx smithers-orchestrator eval --help` for usage details. |
| requires_bin | bunx |
| command | bunx smithers-orchestrator eval |
Run a workflow over a JSON/JSONL eval suite and write a regression report.
| Name | Type | Required | Description |
|---|---|---|---|
workflow | string | yes | Path to a .tsx workflow file |
| Flag | Type | Default | Description |
|---|---|---|---|
--cases | string | JSON or JSONL eval case file | |
--suite | string | Stable suite ID used in run IDs and report paths | |
--runLabel | string | Run label appended to eval run IDs; defaults to current UTC timestamp plus a nonce | |
--dryRun | boolean | false | Plan the suite without launching runs |
--concurrency | number | 1 | Number of eval cases to run at once |
--maxCases | number | Run only the first N cases | |
--report | string | Write report JSON to this path | |
--force | boolean | false | Overwrite an existing eval report |
--includeOutput | boolean | true | Include workflow outputs in the report |
--maxConcurrency | number | Per-workflow max task concurrency | |
--root | string | Tool sandbox root directory | |
--log | boolean | true | Enable NDJSON event log file output |
--logDir | string | NDJSON event logs directory | |
--allowNetwork | boolean | false | Allow bash tool network requests |
--maxOutputBytes | number | Max bytes a single tool call can return | |
--toolTimeoutMs | number | Max wall-clock time per tool call in ms | |
--optimization | string | Apply a Smithers optimization artifact while running the eval suite |