Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/opendcai/dataflow-webui/dataflow-devnpx skills add OpenDCAI/DataFlow-WebUI --skill dataflow-devgit clone --depth 1 https://github.com/OpenDCAI/DataFlow-WebUIWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00095 | $0.03985 |
| Opus 5 | $0.00048 | $0.01992 |
| Sonnet 5 | $0.00019 | $0.00797 |
| Haiku 4.5 | $0.00010 | $0.00398 |
Grade A, and why
dataflow-dev scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 319 lines — stays where its author put it; the contents beside it link to each section on GitHub.
DataFlow 开发助手 (dataflow-dev)
仓库级变更策略（哪些改动被允许）不属于本技能。当你在 DataFlow-WebUI 仓库内工作时， 以该仓库根目录的
CLAUDE.md为唯一真源。
激活时执行的步骤
不要预先加载全部参考文件。 三份参考共约 1400 行；先按意图路由，只读用得上的那一份。
- 探测仓库状态（在 DataFlow 仓库根目录下执行）：
git branch --show-current # 当前分支 git log --oneline -3 # 最近提交 git diff --name-only HEAD~1 HEAD # 最近一次变更文件列表 - 向用户报告当前上下文摘要（1-3行，不要冗长）
- 判断用户意图，按下表只读所需文件，然后进入对应工作流
按需加载表
| 用户意图 | 需要读取 | 不需要读取 |
|---|---|---|
| 新建算子 / Pipeline / Prompt | ${SKILL_DIR}/context/dev_notes.md（规范）；写算子时另加 context/knowledge_base.md 的 §3 算子章节 |
诊断表 |
| 报错 / 诊断 | ${SKILL_DIR}/diagnostics/known_issues.md（先查快速匹配表，命中后只读对应 Issue 小节） |
知识库全文、开发规范 |
| 代码审查 / 规范检查 | ${SKILL_DIR}/context/dev_notes.md |
知识库全文、诊断表 |
| 查 API / 架构问题 | ${SKILL_DIR}/context/knowledge_base.md 中相关章节（按标题定位，不要通读） |
诊断表、开发规范 |
| 知识库更新感知 | 三份都要，因为要逐一比对 | — |
参考文件用途：
context/knowledge_base.md— 架构与 API 参考（约 670 行，按章节查阅）context/dev_notes.md— 开发规范与最佳实践diagnostics/known_issues.md— 已知问题诊断表（Issue #001–#009）
有 MCP 时：算子签名以 get_operator_detail_by_name 为准，它反映实际安装的版本；知识库是静态快照，冲突时以 MCP 为准。
离线模式：以随技能提供的知识库和 core_text 静态参考为准；如果本地 DataFlow 版本不同，使用 Python 的 inspect.signature 做一次本地核对，并说明无法查询实时注册表。
子命令路由
根据用户意图，路由到对应工作流：
| 用户意图关键词 | 执行流程 |
|---|---|
| 新建算子 / new operator / create operator | → 算子创建流程 |
| 新建 Pipeline / new pipeline | → Pipeline 创建流程 |
| 新建 Prompt / new prompt | → Prompt 创建流程 |
| 报错 / error / KeyError / AttributeError / Warning | → 诊断流程 |
| 审查代码 / check / review / 规范检查 | → 规范审查流程 |
| 更新知识库 / sync / check updates / 仓库有新算子 | → 知识库更新感知流程 |
算子创建流程
Step 1: 防重复检查（必须）
在生成代码前，先检查是否已有功能相近算子：
# 查看各模块已注册算子
grep -r "^from \." dataflow/operators/general_text/__init__.py | grep TYPE_CHECKING -A 200 | grep "^ from"
grep -r "^ from" dataflow/operators/text_sft/__init__.py
grep -r "^ from" dataflow/operators/reasoning/__init__.py
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 319 lines · 95 tokens per session scan A 06d517528c41
dataflow-dev is a skill published in the GitHub repository OpenDCAI/DataFlow-WebUI (220 stars, last pushed 6d ago), licensed Apache-2.0. It adds 95 tokens to every session and 3,985 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…