跳到正文
cccskills
免费在 GitHub 公开

reproduce-chat-states

Reproduce chat errors, retries, parallel tool calls, and app creation in a running desktop application. Check how messages and cards appear, including interrupted replies, grouped steps, and error details.

中文简介正在准备中。当前显示英语简介。

原文查看原文说明

Reproduce a known chat state in the running desktop on demand (model provider errors and retries, Code Mode and Den script failures, parallel tool steps, app creation, sequential commands) and check how it renders. Use when fixing or designing chat error, retry, tool-step or app-card UI.

集成工具Slack
查看安装方法

适用场景

  • Check error messages and retry progress
  • Inspect text retained after an interrupted reply
  • Review grouped parallel tool calls
  • Check app cards and chat movement

包含的文件(2)

  • SKILL.md2.6 KB
  • scripts/repro.mjs3.1 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Reproduce chat states

Each state below fires in a new session of the running desktop window through drive-desktop-cdp. Provider states also need fake-model-faults set up.

export CDP_URL=http://127.0.0.1:<cdp port>
R=.opencode/skills/reproduce-chat-states/scripts/repro.mjs
C=.opencode/skills/drive-desktop-cdp/scripts/cdp.mjs
node $R provider-refused && node $C wait-idle && node $C shot /tmp/refused.png

Run node $R with no argument for the list. Pass a second argument (or set GATEWAY_MODEL) to change the real model used for the non-fault states.

States and what to look at

StateFiresCheck
provider-refused, provider-dns, provider-resetengine transport error, retried, then finalone error per turn, plain title, raw text only behind the details icon
provider-retryingsame as reset; watch the first 30s"Reconnecting to the model" with attempt progress, no raw text
provider-midstream, provider-stallcut or stalled replyinterrupted reply keeps partial text
http-429, http-5xx, http-401, http-402HTTP error bodiesthe matching plain-language title and next action
codemode-wrong-tooltools.search(...) in Code Modea recovered step must not read as a hard failure
den-script-rejectedDen script with .thenDen's real reason in details, not a generic placeholder
den-script-service-failureprovider op fails inside a Den script"Script on OpenWork Cloud failed", reason behind the icon
den-script-parallel20 Slack calls in one scriptone readable group, collapsed until opened
parallel-searches10 Exa searches plus Slackclean service names and logos, no routing ids
app-createMCP app build with retriesone card per app that advances through its steps
sequential-commands5 separate bash callsthe chat does not jump (see measure-chat-flicker)

Reading what happened

Stored v2 history shows exactly what the engine recorded (useful when a step looks wrong):

sqlite3 ~/.config/openwork/opencode-v2/state/opencode.db \
  "select type, substr(data,1,400) from session_message where session_id='<ses_…>' order by seq"

For a step inside Code Mode, check state.metadata.toolCalls and state.metadata.openworkMcpResults on the execute part.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Manage OpenWork feature releases with feature flags, switches that control access to new behavior. Register features before coding, release them by organization and deployment, prepare a global shutdown option, and remove temporary flags after rollout.

中文简介正在准备中。当前显示英语简介。

  • Registering a feature before implementation
  • Releasing a feature to selected organizations
  • Separating cloud and self-hosted availability
different-ai/openwork2.4万2026年10月22日 更新

Connect OpenCode browser tools to a locally running OpenWork Electron app to inspect its UI, enter a task, and check the response. This skill supports exploration and debugging through CDP, Chrome's remote control protocol.

中文简介正在准备中。当前显示英语简介。

  • Connecting browser tools to a local Electron development app
  • Inspecting screen state during debugging
  • Entering a task in the composer and running it
different-ai/openwork2.4万2026年10月22日 更新

Check added lines in a change for information identifying customers, prospects, partners, or outside individuals. Report locations and generic remediation without reproducing identities, and escalate cases where the information is already public.

中文简介正在准备中。当前显示英语简介。

  • Checking changes for customer and partner identities
  • Reviewing documents and test data for identifying details
  • Reporting disclosure locations without repeating identities
different-ai/openwork2.4万2026年10月22日 更新

Connects supported AI clients, including Claude Code and Codex, to OpenWork through MCP, a protocol for accessing external tools. It guides account sign-in and verifies that tools for finding and running organization capabilities are available.

中文简介正在准备中。当前显示英语简介。

  • Connecting an AI client to OpenWork organization resources
  • Signing in and verifying available MCP tools
  • Switching organizations or troubleshooting authentication
different-ai/openwork2.4万2026年10月22日 更新

Review an outside contributor’s pull request before running its code. Identify concealed behavior, obfuscated code, risky dependencies, and build hooks, with file locations and specific checks for maintainers to follow up.

中文简介正在准备中。当前显示英语简介。

  • Screening an outside contributor’s pull request before execution
  • Checking encoded code and unexplained network calls
  • Reviewing dependency changes and installation scripts
different-ai/openwork2.4万2026年10月22日 更新

Create an OpenCode plugin for OpenWork with custom tools and hooks, which run code at specific events. The skill builds the plugin file and adds its entry to the project configuration.

中文简介正在准备中。当前显示英语简介。

  • Adding custom tools to OpenWork
  • Creating a plugin to save and list notes
  • Running custom code before or after tool execution
different-ai/openwork2.4万2026年10月22日 更新

查看 different-ai 的全部技能

报告此技能的问题