Skip to content
Agentmart
Submit a tool

workflow

Route a TopoRealm Workflow request to quick, standard, or program and identify whether design, join, or execution is next.

Skill by LUKAWI

IndexedMIT
Write a review

Chalk Labs

Need help deploying workflow?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/lukawi-workflow

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Route a TopoRealm Workflow request to quick, standard, or program and identify whether design, join, or execution is next. Read-only; do not claim or mutate tasks.

Use cases: design, productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/workflow

What's inside

Skill name
workflow
Description
Route a TopoRealm Workflow request to quick, standard, or program and identify whether design, join, or execution is next. Read-only; do not claim or mutate tasks.
License
MIT
Instructions
48 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Workflow Router

只做路由,不点火、不认领、不提交图变更。

1. 多步骤且需要依赖、交接或可追溯验收时使用 Workflow;单点问答或一次性小改直接完成。
2. 关键未知阻止形成可信交付计划时选 `program`;否则一个会话内可完成并验收选 `quick`;其余选 `standard`。
3. 通过 `toporealm cmds` 确认 workflow 模块已装载后,执行只读的 `toporealm wf.next-actions`,读取当前图与调度前沿(ready/frontier/blocked)。用户未指定且存在多图时请用户选择,不改共享 active 图。
4. 未设计或设计未获批准 → `workflow-design`;新会话缺背景 → `workflow-join`;已批准且获执行授权 → `workflow-execute`。

## 共享执行协议

### 唯一操作面

一切读写经 TopoRealm CLI:`toporealm wf.<name> [target] [--input '<json>']`;目录自省
`toporealm cmds --module wf`;读图 `toporealm read/find`;撤销是用户的手
(`toporealm undo/redo`),模块没有 undo。领域命令即写缝——绝不手改图文件绕过门禁。

公开命令(13 个):`wf.create-task`、`wf.create-relation`、`wf.next-actions`、
`wf.transition-task`、`wf.claim-task`、`wf.record-checkpoint`、`wf.record-report`、
`wf.verify-task`、`wf.retry-task`、`wf.activate-fallback`、`wf.record-iteration`、
`wf.set-class`、`wf.assign-domain`。输入 schema 用 `toporealm help wf.<name>` 查看。

### 不变量

- 七态为 `pending → ready → running → passed|failed|blocked`,另有 `cancelled`;running 必须由 claim 进入。
- `wf.depends_on` 的 source 是前置,target 是后继;前置未全部 passed 时不得置 ready 或 claim。
  该门禁由模块 before-commit 钩子对一切前向写入执法(含 CLI 直改);undo/redo 是用户的游标,不受门禁拦截。
- checkpoint 内嵌在任务 `payload.checkpoints`(1.1.0 起无独立 checkpoint 对象,经 `wf.record-checkpoint` 上报);
  execution_report 是独立一等证据对象,写入时自动双写 `wf.report_of` 归属关系(id `rel-of-<报告id>`,
  source=报告 target=任务),payload.taskId 双写冗余。先完成 checkpoint,再写非空 report,再裁决,最后转 passed。
- 任务按领域归类用 `wf.assign-domain`:`wf.domain` 容器对象不存在则自动创建(id = 领域名 slugify,
  title 缺省 = 领域名),并建公共 `member_of` 关系(source=任务 target=领域);重复执行幂等。
- `human` checkpoint 或 verification 的通过只能记录用户真实确认,agent 不得代签(钩子拦截)。
- `independent` 复核默认只是 `reviewSuggested` 建议;不阻塞 self 完成,不得伪造成强制 gate。
- failed/blocked 仅在预算内 retry;预算耗尽后只走已存在的 fallback。iteration 必须用显式 `wf.iterates` 关系留痕。
- `quick` 用于单会话可完成事项,`program` 只在关键未知阻止可信计划时使用,其余为 `standard`;图级档位存于 `wf.settings` 单例。

### 错误处理

- 领域拒绝:命令返回封闭集错误码(`INVALID_INPUT`)或钩子 `VETOED`(消息携带
  `INVALID_TRANSITION` / `DEPENDENCY_UNMET` / `TASK_NOT_COMPLETE` / `HUMAN_CONFIRMATION_REQUIRED` 明细)。
  保留原状态并报告真实原因,不得伪造领域状态。
- `IF_REVISION_MISMATCH`:另一客户端先改了图——重读后重试。
- 结构性错误(悬空边等):core 点名缺失端点并给出 fix 命令,照修即可。

操作前读本协议。回答只给判断、证据、目标图与下一入口,不把建议说成已执行。

From LUKAWI/toporealm-workflow, MIT.

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Sep 29, 2026 (1 month ago)
Last release
Sep 23, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from LUKAWI

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get workflow running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying workflow?

CallGet help
Cite this page

workflow skill. Agent Mart. https://agentmart.dev/skills/lukawi-workflow (last updated 2026-10-24)