OpenClaw

openclaw-flow-kit

Fix common OpenClaw workflow bottlenecks: platform engage-gates/429 backoff helpers (starting with MoltX), standardized JSON result envelopes for chaining scripts, workspace path resolution helpers, and a simple skill release conductor (prepare/publish/draft announcements).

deepseekoracle0 来源安装数☆ 0 仓库 Stars

查看来源 ↗安装与使用 SKILL.md ↗

安装与使用

请根据来源仓库和你的 AI 工具说明安装此技能。先阅读技能文件和依赖要求，再添加到工作流。

SkillHub CLI

skillhub install deepseekoracle/openclaw-flow-kit

打开来源仓库，查看说明和兼容环境后，按来源提供的命令或步骤安装。

将来源链接提供给你的 Agent，让它先读取说明、核对依赖，并向你说明将执行的安装步骤。

查看来源 ↗使用指南

技能说明

OpenClaw Flow Kit

Use this when you hit:

platform engage gates / flaky 429 loops (esp. MoltX)
inconsistent script outputs that make skill-chaining painful
workspace-relative path bugs (writing to skills/state vs state)
repetitive skill release steps (publish + generate announcement drafts)

Quick commands

1) Standardized result envelope for any command

python scripts/run_envelope.py -- cmd /c "echo hello"

Outputs JSON:

ok, exit_code, stdout, stderr, startedAt, endedAt, durationMs

2) MoltX engage-gate helper (read feeds + like/repost)

python scripts/moltx_engage_gate.py --mode minimal

Then run your post normally.

3) Workspace root resolver (import helper)

Use in scripts to find the real workspace root:

from scripts.ws_paths import find_workspace_root
WS = find_workspace_root(__file__)

4) Release conductor (prepare → publish → draft)

python scripts/release_conductor.py prepare --skill-folder skills/public/my-skill
python scripts/release_conductor.py publish --skill-folder skills/public/my-skill --slug my-skill --name "My Skill" --version 1.0.0 --changelog "..."
python scripts/release_conductor.py draft --slug my-skill --name "My Skill" --out tmp/drafts

Notes:

draft generates post text files; it does not post anywhere.

来源信息

Source: github
Repository: openclaw/skills
Path: skills/deepseekoracle/openclaw-flow-kit
Version: —
Commit: 37a7f24ae9428eafb910b18db65631792ee40fb9
分类来源: rules
状态: active
可见性: public

这些信息来自收录时的来源分析，可能滞后；使用前请核对最新来源文件。

来源内容

SKILL.md

# OpenClaw Flow Kit

Use this when you hit:
- platform **engage gates** / flaky **429** loops (esp. MoltX)
- inconsistent script outputs that make skill-chaining painful
- workspace-relative path bugs (writing to skills/state vs state)
- repetitive skill release steps (publish + generate announcement drafts)

## Quick commands

### 1) Standardized result envelope for any command
```bash
python scripts/run_envelope.py -- cmd /c "echo hello"
```
Outputs JSON:
- `ok`, `exit_code`, `stdout`, `stderr`, `startedAt`, `endedAt`, `durationMs`

### 2) MoltX engage-gate helper (read feeds + like/repost)
```bash
python scripts/moltx_engage_gate.py --mode minimal
```
Then run your post normally.

### 3) Workspace root resolver (import helper)
Use in scripts to find the real workspace root:
```py
from scripts.ws_paths import find_workspace_root
WS = find_workspace_root(__file__)
```

### 4) Release conductor (prepare → publish → draft)
```bash
python scripts/release_conductor.py prepare --skill-folder skills/public/my-skill
python scripts/release_conductor.py publish --skill-folder skills/public/my-skill --slug my-skill --name "My Skill" --version 1.0.0 --changelog "..."
python scripts/release_conductor.py draft --slug my-skill --name "My Skill" --out tmp/drafts
```

Notes:
- `draft` generates post text files; it does not post anywhere.

搜索