技能详情（站内镜像，无评论）
作者：Ask Gina @askginadotai
许可证：MIT-0
MIT-0 ·免费使用、修改和重新分发。无需归因。
版本：v1.0.2
统计：⭐ 2 · 492 · 4 current installs · 4 all-time installs
⭐ 2
安装量（当前） 4
🛡 VirusTotal ：可疑 · OpenClaw ：良性
Package：askginadotai/workflows
安全扫描（ClawHub）
- VirusTotal ：可疑
- OpenClaw ：良性
OpenClaw 评估
The skill is an instruction-only playbook for authoring and running sandbox workflows; its requirements and instructions are internally consistent with that purpose.
目的
Name/description (workflow authoring, running, evaluating) match the content: CLI commands, workflow file locations, run artifacts, and evaluation/optimize loop are all directly relevant. No unrelated credentials, binaries, or installs are requested.
说明范围
The SKILL.md and references describe workflow step types that include powerful runtime primitives (exec, fs.promises.*, kv.*, sql, callTool). This is expected for a workflow engine because steps must run shell commands, read/write files, and use KV/SQL. Because the playbook instructs authors/operators to scaffold/open workflows in /workspace/.harness/workflows and to validate/run them, you should review workflow code before executing runs—step…
安装机制
Instruction-only skill with no install spec and no code files to write to disk. Low installation risk.
证书
No environment variables, credentials, or config paths are requested. The playbook explicitly advises least-privilege and not including secrets in logs or skill text.
持久
Skill does not request always:true, does not modify other skills, and is user-invocable only. No long-lived privileges or autonomous persistence are requested.
综合结论
This is a coherent operations playbook for creating and running sandbox workflows. It does describe and rely on workflow step capabilities that can execute shell commands, read and write files, and access KV/SQL; those are normal for a workflow system but mean you should never run unreviewed workflows. Before installing/using: (1) Confirm the 'workflow' CLI/runtime you will call is the expected/trusted tool on your system, (2) review any .ts w…
安装（复制给龙虾 AI）
将下方整段复制到龙虾中文库对话中，由龙虾按 SKILL.md 完成安装。
请把本段交给龙虾中文库（龙虾 AI）执行：为本机安装 OpenClaw 技能「Workflows」。简介：Playbook for authoring, running, evaluating, and improving Gina sandbox workflo…。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装：https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/askginadotai/workflows/SKILL.md
（来源：yingzhi8.cn 技能库）SKILL.md
---
id: workflows
name: "Workflows via Ask Gina"
description: "Playbook for authoring, running, evaluating, and improving Gina sandbox workflows with safe defaults and repeatable operations."
---
# Workflows via Ask Gina Skill
## What It Does
Provides a practical workflow-authoring and operations standard for Gina sandbox automation.
- Creates and validates workflow definitions.
- Runs workflows and inspects artifacts/logs.
- Applies a repeatable eval -> optimize -> compare loop.
- Uses safe TypeScript/SQL/KV patterns for step logic.
## When To Use
- You are creating or maintaining multi-step workflow orchestration.
- You need reproducible debugging from run artifacts.
- You want measurable improvements using baseline comparisons.
## When Not To Use
- The task is a single action with no orchestration requirement.
- You only need high-level strategy language without runnable steps.
- You cannot provide explicit permissions or side effects.
## Inputs
- Workflow intent and success criteria.
- Trigger definition and input schema.
- Required tools/data sources and permission scope.
- Optional baseline run ID for optimization.
## Outputs
- Validated workflow definition (`.ts`).
- Runnable execution with traceable artifacts.
- Evaluation record with baseline comparison.
- Clear rollback path for regressions.
## Core Commands
```bash
workflow create <id>
workflow validate <id>
workflow run <id> [--input JSON]
workflow status <run-id>
workflow logs <run-id> [--step <step-id>]
workflow eval <run-id>
workflow optimize <id> --baseline <run-id>
workflow rollback <id> <opt-run-id>
```
## Setup
1. Confirm workflow tooling is available (`workflow list` should succeed).
2. Scaffold or open the target workflow in `/workspace/.harness/workflows/`.
3. Keep active versions on `@latest.ts` naming when versioned variants exist.
4. Validate before every run: `workflow validate <id>`.
5. For risky changes, capture a baseline run and eval before editing.
## Capability Contract Checklist
For each workflow entry, explicitly define:
- Trigger.
- Inputs.
- Outputs.
- Side effects.
- Failure modes.
- Permission scope.
## Failure Modes
- Validation failure from malformed step definitions.
- Runtime errors in TS/SQL/Bash steps.
- Missing tool permissions or tool availability.
- Data shape changes causing parse/cast failures.
- Timeout/retry exhaustion in external calls.
## Security And Permissions
- Use least privilege by step using `allow` and `block`.
- Declare permissions in the submission contract (no wildcard permissions).
- Never include raw secrets in skill text, logs, or examples.
- Treat writes (files, KV, external posts, trading actions) as explicit side effects.
## Evidence Expectations
- Setup path that a reviewer can execute in under 10 minutes.
- One reproducible run artifact or run log example.
- Clear statement of expected outputs and acceptable failure behavior.
## Optional Directories
```text
workflows/
SKILL.md
references/ # implementation and API details
scripts/ # optional helpers for repeatable checks
assets/ # optional diagrams/screenshots
```
## Reference Material
Deep technical references are intentionally split out:
- `references/cli-and-definition.md`
- `references/eval-optimize-and-artifacts.md`
- `references/polymarket-patterns.md`
Use these as appendices while keeping this file focused on operational usage.