[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M746ZPTAX1Q0FBKAB1CYND7E":3,"skill-same:01M746ZPTAX1Q0FBKAB1CYND7E":96},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":35,"commitSha":36,"skillDirectory":37,"forksCount":38,"repositoryHtmlUrl":39,"repositoryHomepageUrl":40,"ownerHtmlUrl":41,"ownerBlogUrl":42,"ownerDescription":43,"updatedAt":44,"install":45,"resources":57},"01M746ZPTAX1Q0FBKAB1CYND7E","e2e",null,"Writes, runs, and debugs end-to-end tests for web and mobile apps with e2e, the agentic test runner, which mixes agent.act goals with exact locator assertions on a Playwright browser or an iOS or Android device. Also drives a project's running app live over the e2e MCP server to look at a screen, check a UI change, take a screenshot or record a video, and try a locator; and runs bug bashes, parallel e2e explore runs with each bug proven by a repro test. Use when a project depends on e2e or has an e2e.config.ts, when asked for end-to-end, browser, mobile, or agentic UI tests, to check a change in the running app, to QA or bug bash an app or branch, or when an e2e run fails.","hashgraph-online\u002Fawesome-codex-plugins","hashgraph-online","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160371962?v=4","01M6CNB6NRJVNNGN14TR01Q6D3","HOL (Hashgraph Online)","01M6JT1HSKKG9V68KQAQGBA7XP",1575,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: e2e\ndescription: Writes, runs, and debugs end-to-end tests for web and mobile apps with e2e, the agentic test runner, which mixes agent.act goals with exact locator assertions on a Playwright browser or an iOS or Android device. Also drives a project's running app live over the e2e MCP server to look at a screen, check a UI change, take a screenshot or record a video, and try a locator; and runs bug bashes, parallel e2e explore runs with each bug proven by a repro test. Use when a project depends on e2e or has an e2e.config.ts, when asked for end-to-end, browser, mobile, or agentic UI tests, to check a change in the running app, to QA or bug bash an app or branch, or when an e2e run fails.\nmetadata:\n  openclaw:\n    homepage: https:\u002F\u002Fe2e.com\n    requires:\n      bins:\n        - node\n        - npx\n    envVars:\n      - name: AI_GATEWAY_API_KEY\n        required: false\n        description: Model key for agent steps when the project's e2e config uses the Vercel AI Gateway; other providers read their own key.\n---\n\n# e2e: agentic end-to-end tests\n\ne2e runs UI tests that mix agent goals with exact steps. `agent.act` drives\none goal; `agent.assert`, `agent.waitFor`, and `agent.extract` judge the\nscreen; `screen`, `app`, `browser`, and `expect` act and check exactly. UI\ntargets use `@e2e-dev\u002Fweb` for browsers or `@e2e-dev\u002Fmobile` for iOS\nsimulators, Android emulators, and connected phones.\n\n## Start here: pick the reference for the job\n\nRead the one reference that matches the request before doing anything else.\nEach stands on its own; a reference names another topic only for a side\ntrip. `npx e2e guide \u003Ctopic>` prints the installed package's version of a\nreference; prefer it when it differs from this copy.\nThe Rules and Feedback sections below still apply.\n\n| When you are | Topic | Read |\n| --- | --- | --- |\n| Adding e2e to a project, writing `e2e.config.ts`, starting the app from the config, signing in to a model, setting up mobile targets | `setup` | [references\u002Fsetup.md](references\u002Fsetup.md) |\n| Looking at the running app, checking a UI change, taking a screenshot or video, trying a locator, running a test line, or walking a flow while you write its test, over `e2e mcp` | `mcp` | [references\u002Fmcp.md](references\u002Fmcp.md) |\n| Writing or fixing a test: fixtures, locators, actions, matchers, sign-in sessions, the `browser` fixture, API-only tests | `writing-tests` | [references\u002Fwriting-tests.md](references\u002Fwriting-tests.md) |\n| Adding `agent.*` steps, picking a model, cutting cost, budgets, the replay cache | `agent` | [references\u002Fagent.md](references\u002Fagent.md) |\n| Running tests: CLI flags, CI, reporters, trace pages, `report.json` for scripts, exit codes | `running` | [references\u002Frunning.md](references\u002Frunning.md) |\n| Handling a failed run: error codes and their fixes, `--headed`, `--debug`, `--ai-trace` | `debugging` | [references\u002Fdebugging.md](references\u002Fdebugging.md) |\n| Exploring an app toward a goal without a test file: `e2e explore`, its budgets and verdict | `explore` | [references\u002Fexplore.md](references\u002Fexplore.md) |\n| Asked to bug bash, QA, or hunt for bugs across an app or a branch | `bug-bash` | [references\u002Fbug-bash.md](references\u002Fbug-bash.md) |\n\nFor anything the references do not cover, the full documentation ships in\nthe `docs\u002F` directory of the installed `e2e` package\n(`node_modules\u002Fe2e\u002Fdocs` in a single-package project); a link such as\n`\u002Freference\u002Fcli` is `docs\u002Freference\u002Fcli.mdx`. Complete projects for Vite,\nNext.js, Astro, Angular, Expo, SwiftUI, Jetpack Compose, Kotlin Multiplatform,\nFlutter, and .NET MAUI are in https:\u002F\u002Fgithub.com\u002Fe2e-inc\u002Fe2e\u002Ftree\u002Fmain\u002Fexamples.\n\n## Quick start\n\nA config with one web target and a model, and a test that drives one goal\nand pins the outcome:\n\n```ts\n\u002F\u002F e2e.config.ts\nimport type { E2EConfig } from 'e2e';\nimport { web } from '@e2e-dev\u002Fweb';\nimport { gateway } from 'ai';\n\nexport default {\n  targets: [\n    {\n      engine: web(),\n      app: {\n        url: 'http:\u002F\u002F127.0.0.1:3000',\n        command: { executable: 'pnpm', args: ['dev'], log: '.e2e\u002Flogs\u002Fapp.log' },\n      },\n    },\n  ],\n  \u002F\u002F The model behind every agent.* step: an AI SDK instance; gateway() from 'ai' reads AI_GATEWAY_API_KEY or a Vercel OIDC token.\n  agents: {\n    default: {\n      model: gateway('openai\u002Fgpt-6-luna-fast'),\n      system: 'You are a thorough QA agent. Verify every outcome on screen.',\n    },\n  },\n} satisfies E2EConfig;\n```\n\n```ts\n\u002F\u002F tests\u002Fbilling.e2e.ts\nimport { describe, test } from '@e2e-dev\u002Fweb';\nimport { expect } from 'e2e';\n\ndescribe('billing', () => {\n  test('a member upgrades to Pro', async ({ app, agent, screen, browser }) => {\n    await app.open('\u002Fsettings\u002Fbilling');\n\n    await agent.act('upgrade the workspace to the Pro plan');\n\n    await expect(screen.getByRole('status')).toContainText('Pro');\n    await expect(browser).toHaveURL('\u002Fsettings\u002Fbilling');\n  });\n});\n```\n\n## Workflow: writing a test\n\n1. Look at what exists: `e2e.config.ts`, `.mts`, `.js`, or `.mjs`, the `tests` glob\n   (default `tests\u002F**\u002F*.e2e.ts`), `e2e` in `package.json`. Nothing there:\n   follow `setup` when the user has requested e2e (see Rules).\n2. With a connected `e2e mcp` server (topic `mcp`), write the test while\n   you walk the flow, not from memory after the walk. A test written\n   afterwards fails on each name or step you got wrong, one run at a time.\n   Before `open_session`, create `tests\u002F\u003Cfeature>.e2e.ts`, opening on the\n   screen it covers (step 3). Then add each step to the file as you take it\n   in the session. An `act` goal goes in before you walk its path, and the\n   path's actions stay out of the test. A `screen` line is the `Test line:`\n   its action printed, copied as is. When the outcome shows, run its\n   `expect` with `run_code` and add it once it passes; a passing statement\n   also joins `test_lines`; checks go in the test's assert block at the end.\n   The walk ends with the test written. Without a\n   server, learn the accessible names the app renders from the components or\n   a `--headed` run.\n3. Shape the file before the steps (topic `writing-tests`, \"Structure\"):\n   one top-level `describe` per feature, and in it one test per behavior,\n   titled with the outcome it verifies. Each test runs arrange, act, assert,\n   in that order: reach the starting state, do the one thing under test,\n   then every `expect` or `agent.assert` at the end. A second outcome worth\n   checking is a second test, so a failing title says what broke.\n   A step is an `agent.act` goal or `screen` lines. Drive the flow with\n   `agent.act`, one atomic goal per call; three goals make three calls.\n   Start the test on the screen it covers, since every rerun pays\n   for the steps before: in a browser a saved session instead of the\n   sign-in form and `app.open` on that screen's path, on a device\n   `device.openLink`, and data from the app's API (topic `writing-tests`).\n   Exact values go through `screen`: a sign-in form in a setup test, a field\n   that must receive one specific string, a count that must be one number.\n4. Run only the test you are working on:\n   `npx e2e run tests\u002F\u003Cfeature>.e2e.ts:\u003Cline>` (over MCP,\n   `run_tests {files: [\"tests\u002F\u003Cfeature>.e2e.ts:\u003Cline>\"]}`, the flags as\n   camelCase fields such as `retries: 0`), the line its `test(` opens on,\n   with the flags in \"Rerunning a test while you fix it\" below. Agent\n   steps need a model in the config and that provider's authentication (a\n   saved subscription login, an API key); a local endpoint may need none.\n   Tests without agent steps need no model.\n5. When it fails, fix it from what the run recorded (next section).\n\n## Workflow: fixing a failed test\n\nThe run already saw every screen up to the failure. Fix from what it\nrecorded; never walk the flow again from the start over MCP just to see a\nscreen.\n\n1. Read the failure block in the terminal: the failing line with its code\n   frame, the message (an assertion's `expected` and `observed`), where the\n   screen was (`❯ at`: a URL on the web, the app and screen title on a\n   device), and the nodes nearest a failed locator (`❯ on screen`). Then\n   read the trace page it names (`❯ trace .e2e\u002Fresults\u002F\u003Ctest>\u002Ftrace.md`):\n   every step with what it did, what the app logged, the agent's turns, and\n   `## Screen at failure`, the screen one node per line (topic `debugging`).\n2. Fix every line that evidence covers, not only the one that failed: the\n   screen at failure shows the names the rest of that screen's lines need.\n   Fix the locator, the expectation, or the app. Never add a sleep or a\n   fixed delay (`setTimeout`, `waitForTimeout`): wait for the state you\n   need with `agent.waitFor('\u003Cthat state>')` instead (topic `agent`; it\n   needs a model).\n3. A test stops at its first failed step (an `expect.soft` lets it go on),\n   and every wrong line after it costs another run. Before rerunning, check\n   the lines after it as far as you can: against the screen at failure, the\n   app source, or a live session.\n4. For a screen the evidence and the source leave out, look live (topic\n   `mcp`). `open_session {test: \"tests\u002F\u003Cfeature>.e2e.ts\"}` (with `:\u003Cline>`\n   of its `test(` call when the file holds several) runs the test and stops\n   on the live screen where it failed. Act out the lines after it there and\n   put the `Test line:` each action prints in place of a wrong one, so the\n   next rerun catches every mistake. Otherwise open or reuse a session, go\n   straight to that screen, then `observe` and `locate`, or `run_code` for\n   the fixed line and the ones after it. On the web, `navigate` to the path\n   of the `❯ at` URL or the screen's route, and tap only to reach a state no\n   URL opens. A device session starts on the app's launch screen: take the\n   shortest path from there.\n5. Rerun only that test, the way the next section says, or in a new\n   session with `open_session {test}` (topic `mcp` says when a session\n   must close first). Repeat with the next failure.\n\n## Rerunning a test while you fix it\n\nEvery run of a failing test waits out its failure, and every `agent.act`\nthat runs live spends model calls. Keep each rerun to the one test and the\nlines you changed:\n\n- Run by `file:line`, not the file or the suite. After an edit above the\n  test, look its `test(` line up again: a stale line runs another test, or\n  fails with `NO_TESTS` listing the lines tests start at.\n- Add `--action-timeout 5000`: a locator that matches nothing fails in 5 s\n  instead of the default 30 s. It also bounds each action inside an agent\n  step, so keep it longer than the slowest screen takes to load. A `screen`\n  call's own `{ timeout }` still wins. An `expect` already gives up after\n  5 s by default (`assertionTimeout`).\n- Keep the replay cache on. A failed run keeps the recording of every goal\n  a later check or wait confirmed (the wait closing the arrange block, topic\n  `writing-tests`), and the next run replays those with no model call.\n  New, changed, and failed goals run live, and so does a replay the trace\n  page shows as handed off. `agent.assert`, `agent.waitFor`, and\n  `agent.extract` always call the model. Leave `--no-cache` off: it sends every goal to the model\n  again. Only a `read-write` cache records: the local default, but not with\n  `CI` set and no cache mode in the config, nor in strict mode. A stopped\n  run (Ctrl-C, a command your tool's timeout kills, `--max-duration`) saves\n  nothing, so give the command a timeout longer than the test.\n- Don't rename the test, its `describe`, or its file, and don't reword a\n  goal, change its params, or change the agent's `context` or the test's\n  `agentContext` once a goal passes. Each is part of a recording's key, so\n  every goal it touches records again live. Fixing a locator or an\n  expectation keeps the recordings.\n- Add `--retries 0` when retries are on (`retries` in the config, or `CI`\n  set, where the default is 1); a `retries` option on the test or its\n  `describe` wins over the flag. Outside strict mode (`--strict-cache`, `cache.strict`), a\n  retry never replays: it runs every goal live and waits out the failure\n  again.\n- Once the test passes, drop `--action-timeout` and run it one time with\n  `--no-cache`: every goal runs live at the config's timeouts, which shows\n  the agent still reaches it without its recordings.\n\n## Rules\n\n- In a project without e2e, set it up when the user requests it. Otherwise\n  ask before adding a test framework, and name any existing framework.\n- Live MCP tools need an `e2e mcp` server in the client: the e2e plugin's, or\n  one registered in the project (topic `mcp`). In a directory with no e2e config, a live look still\n  works: `open_session {engine: \"@e2e-dev\u002Fweb\", app: {url: \"http:\u002F\u002Flocalhost:3000\"}}` opens a session\n  on the engine's defaults. Tests need a config. Tests run with the CLI, or with the server's\n  `run_tests`, the same run without a shell.\n  Keep a session open across runs while you write or fix a web test; topic\n  `mcp` says when a run can go beside it, and what to do on\n  `APP_ALREADY_RUNNING`.\n- Run the CLI as `npx e2e ...` (or `pnpm exec e2e ...`).\n- The config is `export default { ... } satisfies E2EConfig` with\n  `import type { E2EConfig } from 'e2e'`. `targets` is required; a UI target\n  names an engine and declares the app beside it: `{ engine: web(), app: { url, command } }`.\n  A tools-only target can omit the engine and set `platform`.\n- Import `test`, `describe`, the hooks, `expect`, `credentials`, and\n  `secrets` from `e2e`. A test that uses the `browser` fixture imports `test`,\n  `describe`, and the hooks from `@e2e-dev\u002Fweb`: the same runtime functions,\n  typed with `browser`.\n- Config and tests are ES modules whatever `package.json` sets as `type`.\n- Locators resolve when used. Actions wait for readiness and `expect` retries\n  assertions. Reads such as `textContent()` wait for one match until the\n  action timeout, then answer the current value; `count()` and `isVisible()`\n  answer from the current screen. Nothing waits for a value to change, so use\n  a matcher when a value has to settle.\n- A locator that matches two nodes fails with `LOCATOR_AMBIGUOUS`; narrow it\n  (topic `writing-tests`).\n- Secrets never appear in test code. Declare accounts under `credentials` and\n  every other sensitive value under `secrets` in the config; resolve with\n  `credentials.user(name).password` or `secrets.get(name)` (separate\n  namespaces: `secrets.get` never returns a password), and hand the opaque\n  `Secret` only to `fill()` or `agent.act` params.\n- Tests: one top-level `describe` per feature, one behavior per test,\n  arrange, act, assert with every assertion last (topic `writing-tests`).\n  Bad: one test that adds a todo, checks it, marks it done, and checks\n  again. Good: `'adds a todo to the list'` and `'marks a todo as done'`,\n  the second adding its todo in its arrange block.\n- Agent instructions: one goal per `act`, the wording on screen, real values\n  in params. A goal is atomic: one outcome that one check confirms. Two\n  outcomes are two goals in two `act` calls (topic `agent`), never `'add a\n  todo and mark it done'`. Judge meaning, not phrasing: `toContain('Pro')`,\n  not an exact sentence a model produced.\n- End every test that runs an agent goal with a verifying check: a\n  locator matcher (`expect(locator)`), an engine matcher\n  (`expect(browser).toHaveURL`), or `agent.assert`. A passing one confirms\n  every `act` before it, so they can be cached and replayed without model\n  calls, and so do `locator.waitFor`, `browser.waitForURL`, and\n  `agent.waitFor`. Reads, `expect.poll`, and plain-value `expect` confirm\n  nothing (topic `agent`, \"The replay cache\").\n- Shape the agent for this app: `context` for vocabulary the screens use,\n  `system` for how it works, tools for a test API, named personas under\n  `agents`. When a step fails, tighten the goal first, then the context, then\n  the agent.\n- `.e2e\u002F` is output (`report.json`, `results\u002F`, `cache\u002F`, `logs\u002F`; the\n  config's `output` moves the report and artifacts, never `cache\u002F` or the\n  app's log). Read it, never edit it.\n\n## Feedback\n\nWhen e2e itself gets in your way, offer to tell the e2e team: a command or\nAPI that broke (`bug`), docs or this skill that misled you (`docs`), or a\ncapability you needed and did not find (`feature`). Once per problem, after\nyou worked around it or gave up. Never report:\n\n- failures of the app under test, or its own behavior;\n- a problem you saw once and could not see again;\n- anything on an outdated e2e: the command refuses it, so upgrade\n  (`npx e2e@latest upgrade`) and check the problem still happens first.\n\nSearch [existing issues](https:\u002F\u002Fgithub.com\u002Fe2e-inc\u002Fe2e\u002Fissues) first\n(`gh issue list -R e2e-inc\u002Fe2e --search \"\u003Cwords>\"`); comment on a match\ninstead of sending a new report.\n\nA `bug` needs `--repro`, steps that reproduce it every time without app\ncontent. Without such steps, or when the bug needs your app to show, omit\n`--repro`: the command then prints a prefilled GitHub issue and sends\nnothing; show the user the link and tell them to add a minimal reproduction\nand submit it. Never file the issue yourself.\n\nRun it with `--dry-run` first, show the user what would be sent, and send\nit only when they agree.\n\n```bash\nnpx e2e feedback --type bug -m \"\u003Cone or two sentences>\" \\\n  --repro \"\u003Csteps that reproduce it every time>\" --error-code LOCATOR_AMBIGUOUS \\\n  --task \"\u003Cwhat you were doing>\" --expected \"\u003C...>\" --actual \"\u003Cerror message>\" \\\n  --approach \"\u003Cwhat you tried>\" --command \"\u003Ce2e command>\" --agent \"\u003Cagent \u002F model>\"\n```\n\nDescribe e2e's behavior only, briefly: never paste app content, page text,\ntest files, URLs of private apps, or credentials. Secret-named environment\nvariables and common token shapes are redacted, but do not rely on it. After\nsending, give the user the reference id it prints.\n","# e2e: agentic end-to-end tests\n\ne2e runs UI tests that mix agent goals with exact steps. `agent.act` drives\none goal; `agent.assert`, `agent.waitFor`, and `agent.extract` judge the\nscreen; `screen`, `app`, `browser`, and `expect` act and check exactly. UI\ntargets use `@e2e-dev\u002Fweb` for browsers or `@e2e-dev\u002Fmobile` for iOS\nsimulators, Android emulators, and connected phones.\n\n## Start here: pick the reference for the job\n\nRead the one reference that matches the request before doing anything else.\nEach stands on its own; a reference names another topic only for a side\ntrip. `npx e2e guide \u003Ctopic>` prints the installed package's version of a\nreference; prefer it when it differs from this copy.\nThe Rules and Feedback sections below still apply.\n\n| When you are | Topic | Read |\n| --- | --- | --- |\n| Adding e2e to a project, writing `e2e.config.ts`, starting the app from the config, signing in to a model, setting up mobile targets | `setup` | [references\u002Fsetup.md](references\u002Fsetup.md) |\n| Looking at the running app, checking a UI change, taking a screenshot or video, trying a locator, running a test line, or walking a flow while you write its test, over `e2e mcp` | `mcp` | [references\u002Fmcp.md](references\u002Fmcp.md) |\n| Writing or fixing a test: fixtures, locators, actions, matchers, sign-in sessions, the `browser` fixture, API-only tests | `writing-tests` | [references\u002Fwriting-tests.md](references\u002Fwriting-tests.md) |\n| Adding `agent.*` steps, picking a model, cutting cost, budgets, the replay cache | `agent` | [references\u002Fagent.md](references\u002Fagent.md) |\n| Running tests: CLI flags, CI, reporters, trace pages, `report.json` for scripts, exit codes | `running` | [references\u002Frunning.md](references\u002Frunning.md) |\n| Handling a failed run: error codes and their fixes, `--headed`, `--debug`, `--ai-trace` | `debugging` | [references\u002Fdebugging.md](references\u002Fdebugging.md) |\n| Exploring an app toward a goal without a test file: `e2e explore`, its budgets and verdict | `explore` | [references\u002Fexplore.md](references\u002Fexplore.md) |\n| Asked to bug bash, QA, or hunt for bugs across an app or a branch | `bug-bash` | [references\u002Fbug-bash.md](references\u002Fbug-bash.md) |\n\nFor anything the references do not cover, the full documentation ships in\nthe `docs\u002F` directory of the installed `e2e` package\n(`node_modules\u002Fe2e\u002Fdocs` in a single-package project); a link such as\n`\u002Freference\u002Fcli` is `docs\u002Freference\u002Fcli.mdx`. Complete projects for Vite,\nNext.js, Astro, Angular, Expo, SwiftUI, Jetpack Compose, Kotlin Multiplatform,\nFlutter, and .NET MAUI are in https:\u002F\u002Fgithub.com\u002Fe2e-inc\u002Fe2e\u002Ftree\u002Fmain\u002Fexamples.\n\n## Quick start\n\nA config with one web target and a model, and a test that drives one goal\nand pins the outcome:\n\n```ts\n\u002F\u002F e2e.config.ts\nimport type { E2EConfig } from 'e2e';\nimport { web } from '@e2e-dev\u002Fweb';\nimport { gateway } from 'ai';\n\nexport default {\n  targets: [\n    {\n      engine: web(),\n      app: {\n        url: 'http:\u002F\u002F127.0.0.1:3000',\n        command: { executable: 'pnpm', args: ['dev'], log: '.e2e\u002Flogs\u002Fapp.log' },\n      },\n    },\n  ],\n  \u002F\u002F The model behind every agent.* step: an AI SDK instance; gateway() from 'ai' reads AI_GATEWAY_API_KEY or a Vercel OIDC token.\n  agents: {\n    default: {\n      model: gateway('openai\u002Fgpt-6-luna-fast'),\n      system: 'You are a thorough QA agent. Verify every outcome on screen.',\n    },\n  },\n} satisfies E2EConfig;\n```\n\n```ts\n\u002F\u002F tests\u002Fbilling.e2e.ts\nimport { describe, test } from '@e2e-dev\u002Fweb';\nimport { expect } from 'e2e';\n\ndescribe('billing', () => {\n  test('a member upgrades to Pro', async ({ app, agent, screen, browser }) => {\n    await app.open('\u002Fsettings\u002Fbilling');\n\n    await agent.act('upgrade the workspace to the Pro plan');\n\n    await expect(screen.getByRole('status')).toContainText('Pro');\n    await expect(browser).toHaveURL('\u002Fsettings\u002Fbilling');\n  });\n});\n```\n\n## Workflow: writing a test\n\n1. Look at what exists: `e2e.config.ts`, `.mts`, `.js`, or `.mjs`, the `tests` glob\n   (default `tests\u002F**\u002F*.e2e.ts`), `e2e` in `package.json`. Nothing there:\n   follow `setup` when the user has requested e2e (see Rules).\n2. With a connected `e2e mcp` server (topic `mcp`), write the test while\n   you walk the flow, not from memory after the walk. A test written\n   afterwards fails on each name or step you got wrong, one run at a time.\n   Before `open_session`, create `tests\u002F\u003Cfeature>.e2e.ts`, opening on the\n   screen it covers (step 3). Then add each step to the file as you take it\n   in the session. An `act` goal goes in before you walk its path, and the\n   path's actions stay out of the test. A `screen` line is the `Test line:`\n   its action printed, copied as is. When the outcome shows, run its\n   `expect` with `run_code` and add it once it passes; a passing statement\n   also joins `test_lines`; checks go in the test's assert block at the end.\n   The walk ends with the test written. Without a\n   server, learn the accessible names the app renders from the components or\n   a `--headed` run.\n3. Shape the file before the steps (topic `writing-tests`, \"Structure\"):\n   one top-level `describe` per feature, and in it one test per behavior,\n   titled with the outcome it verifies. Each test runs arrange, act, assert,\n   in that order: reach the starting state, do the one thing under test,\n   then every `expect` or `agent.assert` at the end. A second outcome worth\n   checking is a second test, so a failing title says what broke.\n   A step is an `agent.act` goal or `screen` lines. Drive the flow with\n   `agent.act`, one atomic goal per call; three goals make three calls.\n   Start the test on the screen it covers, since every rerun pays\n   for the steps before: in a browser a saved session instead of the\n   sign-in form and `app.open` on that screen's path, on a device\n   `device.openLink`, and data from the app's API (topic `writing-tests`).\n   Exact values go through `screen`: a sign-in form in a setup test, a field\n   that must receive one specific string, a count that must be one number.\n4. Run only the test you are working on:\n   `npx e2e run tests\u002F\u003Cfeature>.e2e.ts:\u003Cline>` (over MCP,\n   `run_tests {files: [\"tests\u002F\u003Cfeature>.e2e.ts:\u003Cline>\"]}`, the flags as\n   camelCase fields such as `retries: 0`), the line its `test(` opens on,\n   with the flags in \"Rerunning a test while you fix it\" below. Agent\n   steps need a model in the config and that provider's authentication (a\n   saved subscription login, an API key); a local endpoint may need none.\n   Tests without agent steps need no model.\n5. When it fails, fix it from what the run recorded (next section).\n\n## Workflow: fixing a failed test\n\nThe run already saw every screen up to the failure. Fix from what it\nrecorded; never walk the flow again from the start over MCP just to see a\nscreen.\n\n1. Read the failure block in the terminal: the failing line with its code\n   frame, the message (an assertion's `expected` and `observed`), where the\n   screen was (`❯ at`: a URL on the web, the app and screen title on a\n   device), and the nodes nearest a failed locator (`❯ on screen`). Then\n   read the trace page it names (`❯ trace .e2e\u002Fresults\u002F\u003Ctest>\u002Ftrace.md`):\n   every step with what it did, what the app logged, the agent's turns, and\n   `## Screen at failure`, the screen one node per line (topic `debugging`).\n2. Fix every line that evidence covers, not only the one that failed: the\n   screen at failure shows the names the rest of that screen's lines need.\n   Fix the locator, the expectation, or the app. Never add a sleep or a\n   fixed delay (`setTimeout`, `waitForTimeout`): wait for the state you\n   need with `agent.waitFor('\u003Cthat state>')` instead (topic `agent`; it\n   needs a model).\n3. A test stops at its first failed step (an `expect.soft` lets it go on),\n   and every wrong line after it costs another run. Before rerunning, check\n   the lines after it as far as you can: against the screen at failure, the\n   app source, or a live session.\n4. For a screen the evidence and the source leave out, look live (topic\n   `mcp`). `open_session {test: \"tests\u002F\u003Cfeature>.e2e.ts\"}` (with `:\u003Cline>`\n   of its `test(` call when the file holds several) runs the test and stops\n   on the live screen where it failed. Act out the lines after it there and\n   put the `Test line:` each action prints in place of a wrong one, so the\n   next rerun catches every mistake. Otherwise open or reuse a session, go\n   straight to that screen, then `observe` and `locate`, or `run_code` for\n   the fixed line and the ones after it. On the web, `navigate` to the path\n   of the `❯ at` URL or the screen's route, and tap only to reach a state no\n   URL opens. A device session starts on the app's launch screen: take the\n   shortest path from there.\n5. Rerun only that test, the way the next section says, or in a new\n   session with `open_session {test}` (topic `mcp` says when a session\n   must close first). Repeat with the next failure.\n\n## Rerunning a test while you fix it\n\nEvery run of a failing test waits out its failure, and every `agent.act`\nthat runs live spends model calls. Keep each rerun to the one test and the\nlines you changed:\n\n- Run by `file:line`, not the file or the suite. After an edit above the\n  test, look its `test(` line up again: a stale line runs another test, or\n  fails with `NO_TESTS` listing the lines tests start at.\n- Add `--action-timeout 5000`: a locator that matches nothing fails in 5 s\n  instead of the default 30 s. It also bounds each action inside an agent\n  step, so keep it longer than the slowest screen takes to load. A `screen`\n  call's own `{ timeout }` still wins. An `expect` already gives up after\n  5 s by default (`assertionTimeout`).\n- Keep the replay cache on. A failed run keeps the recording of every goal\n  a later check or wait confirmed (the wait closing the arrange block, topic\n  `writing-tests`), and the next run replays those with no model call.\n  New, changed, and failed goals run live, and so does a replay the trace\n  page shows as handed off. `agent.assert`, `agent.waitFor`, and\n  `agent.extract` always call the model. Leave `--no-cache` off: it sends every goal to the model\n  again. Only a `read-write` cache records: the local default, but not with\n  `CI` set and no cache mode in the config, nor in strict mode. A stopped\n  run (Ctrl-C, a command your tool's timeout kills, `--max-duration`) saves\n  nothing, so give the command a timeout longer than the test.\n- Don't rename the test, its `describe`, or its file, and don't reword a\n  goal, change its params, or change the agent's `context` or the test's\n  `agentContext` once a goal passes. Each is part of a recording's key, so\n  every goal it touches records again live. Fixing a locator or an\n  expectation keeps the recordings.\n- Add `--retries 0` when retries are on (`retries` in the config, or `CI`\n  set, where the default is 1); a `retries` option on the test or its\n  `describe` wins over the flag. Outside strict mode (`--strict-cache`, `cache.strict`), a\n  retry never replays: it runs every goal live and waits out the failure\n  again.\n- Once the test passes, drop `--action-timeout` and run it one time with\n  `--no-cache`: every goal runs live at the config's timeouts, which shows\n  the agent still reaches it without its recordings.\n\n## Rules\n\n- In a project without e2e, set it up when the user requests it. Otherwise\n  ask before adding a test framework, and name any existing framework.\n- Live MCP tools need an `e2e mcp` server in the client: the e2e plugin's, or\n  one registered in the project (topic `mcp`). In a directory with no e2e config, a live look still\n  works: `open_session {engine: \"@e2e-dev\u002Fweb\", app: {url: \"http:\u002F\u002Flocalhost:3000\"}}` opens a session\n  on the engine's defaults. Tests need a config. Tests run with the CLI, or with the server's\n  `run_tests`, the same run without a shell.\n  Keep a session open across runs while you write or fix a web test; topic\n  `mcp` says when a run can go beside it, and what to do on\n  `APP_ALREADY_RUNNING`.\n- Run the CLI as `npx e2e ...` (or `pnpm exec e2e ...`).\n- The config is `export default { ... } satisfies E2EConfig` with\n  `import type { E2EConfig } from 'e2e'`. `targets` is required; a UI target\n  names an engine and declares the app beside it: `{ engine: web(), app: { url, command } }`.\n  A tools-only target can omit the engine and set `platform`.\n- Import `test`, `describe`, the hooks, `expect`, `credentials`, and\n  `secrets` from `e2e`. A test that uses the `browser` fixture imports `test`,\n  `describe`, and the hooks from `@e2e-dev\u002Fweb`: the same runtime functions,\n  typed with `browser`.\n- Config and tests are ES modules whatever `package.json` sets as `type`.\n- Locators resolve when used. Actions wait for readiness and `expect` retries\n  assertions. Reads such as `textContent()` wait for one match until the\n  action timeout, then answer the current value; `count()` and `isVisible()`\n  answer from the current screen. Nothing waits for a value to change, so use\n  a matcher when a value has to settle.\n- A locator that matches two nodes fails with `LOCATOR_AMBIGUOUS`; narrow it\n  (topic `writing-tests`).\n- Secrets never appear in test code. Declare accounts under `credentials` and\n  every other sensitive value under `secrets` in the config; resolve with\n  `credentials.user(name).password` or `secrets.get(name)` (separate\n  namespaces: `secrets.get` never returns a password), and hand the opaque\n  `Secret` only to `fill()` or `agent.act` params.\n- Tests: one top-level `describe` per feature, one behavior per test,\n  arrange, act, assert with every assertion last (topic `writing-tests`).\n  Bad: one test that adds a todo, checks it, marks it done, and checks\n  again. Good: `'adds a todo to the list'` and `'marks a todo as done'`,\n  the second adding its todo in its arrange block.\n- Agent instructions: one goal per `act`, the wording on screen, real values\n  in params. A goal is atomic: one outcome that one check confirms. Two\n  outcomes are two goals in two `act` calls (topic `agent`), never `'add a\n  todo and mark it done'`. Judge meaning, not phrasing: `toContain('Pro')`,\n  not an exact sentence a model produced.\n- End every test that runs an agent goal with a verifying check: a\n  locator matcher (`expect(locator)`), an engine matcher\n  (`expect(browser).toHaveURL`), or `agent.assert`. A passing one confirms\n  every `act` before it, so they can be cached and replayed without model\n  calls, and so do `locator.waitFor`, `browser.waitForURL`, and\n  `agent.waitFor`. Reads, `expect.poll`, and plain-value `expect` confirm\n  nothing (topic `agent`, \"The replay cache\").\n- Shape the agent for this app: `context` for vocabulary the screens use,\n  `system` for how it works, tools for a test API, named personas under\n  `agents`. When a step fails, tighten the goal first, then the context, then\n  the agent.\n- `.e2e\u002F` is output (`report.json`, `results\u002F`, `cache\u002F`, `logs\u002F`; the\n  config's `output` moves the report and artifacts, never `cache\u002F` or the\n  app's log). Read it, never edit it.\n\n## Feedback\n\nWhen e2e itself gets in your way, offer to tell the e2e team: a command or\nAPI that broke (`bug`), docs or this skill that misled you (`docs`), or a\ncapability you needed and did not find (`feature`). Once per problem, after\nyou worked around it or gave up. Never report:\n\n- failures of the app under test, or its own behavior;\n- a problem you saw once and could not see again;\n- anything on an outdated e2e: the command refuses it, so upgrade\n  (`npx e2e@latest upgrade`) and check the problem still happens first.\n\nSearch [existing issues](https:\u002F\u002Fgithub.com\u002Fe2e-inc\u002Fe2e\u002Fissues) first\n(`gh issue list -R e2e-inc\u002Fe2e --search \"\u003Cwords>\"`); comment on a match\ninstead of sending a new report.\n\nA `bug` needs `--repro`, steps that reproduce it every time without app\ncontent. Without such steps, or when the bug needs your app to show, omit\n`--repro`: the command then prints a prefilled GitHub issue and sends\nnothing; show the user the link and tell them to add a minimal reproduction\nand submit it. Never file the issue yourself.\n\nRun it with `--dry-run` first, show the user what would be sent, and send\nit only when they agree.\n\n```bash\nnpx e2e feedback --type bug -m \"\u003Cone or two sentences>\" \\\n  --repro \"\u003Csteps that reproduce it every time>\" --error-code LOCATOR_AMBIGUOUS \\\n  --task \"\u003Cwhat you were doing>\" --expected \"\u003C...>\" --actual \"\u003Cerror message>\" \\\n  --approach \"\u003Cwhat you tried>\" --command \"\u003Ce2e command>\" --agent \"\u003Cagent \u002F model>\"\n```\n\nDescribe e2e's behavior only, briefly: never paste app content, page text,\ntest files, URLs of private apps, or credentials. Secret-named environment\nvariables and common token shapes are redacted, but do not rely on it. After\nsending, give the user the reference id it prints.\n",{"name":5,"metadata":24,"description":7},{"openclaw":25},{"envVars":26,"homepage":30,"requires":31},[27],{"name":28,"required":20,"description":29},"AI_GATEWAY_API_KEY","Model key for agent steps when the project's e2e config uses the Vercel AI Gateway; other providers read their own key.","https:\u002F\u002Fe2e.com",{"bins":32},[33,34],"node","npx","main","2a6d24be2ed571c5ff2319c506b60741bbcb1cb0","plugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e",491,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins","https:\u002F\u002Fhol.org\u002Fplugins\u002Fbest-codex-plugins","https:\u002F\u002Fgithub.com\u002Fhashgraph-online","https:\u002F\u002Fhol.org","Open-source standards and developer tools for interoperable AI agents: discovery, identity, communication, payments, privacy and security.","2026-11-14 14:06:13",{"source":8,"skillName":5,"commands":46,"prompt":55,"zipAvailable":56},[47,49,52],{"tool":34,"command":48},"npx skills add hashgraph-online\u002Fawesome-codex-plugins --skill e2e",{"tool":50,"command":51},"bunx","bunx skills add hashgraph-online\u002Fawesome-codex-plugins --skill e2e",{"tool":53,"command":54},"pnpm","pnpm dlx skills add hashgraph-online\u002Fawesome-codex-plugins --skill e2e","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002FSKILL.md",true,[58,64,68,72,76,80,84,88,92],{"path":59,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":62,"sourceHtmlUrl":63},"agents\u002Fopenai.yaml","REFERENCE","参考文档",375,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002Fagents\u002Fopenai.yaml",{"path":65,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":66,"sourceHtmlUrl":67},"references\u002Fagent.md",25087,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002Freferences\u002Fagent.md",{"path":69,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":70,"sourceHtmlUrl":71},"references\u002Fbug-bash.md",16804,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002Freferences\u002Fbug-bash.md",{"path":73,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":74,"sourceHtmlUrl":75},"references\u002Fdebugging.md",16482,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002Freferences\u002Fdebugging.md",{"path":77,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":78,"sourceHtmlUrl":79},"references\u002Fexplore.md",8545,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002Freferences\u002Fexplore.md",{"path":81,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":82,"sourceHtmlUrl":83},"references\u002Fmcp.md",27162,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002Freferences\u002Fmcp.md",{"path":85,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":86,"sourceHtmlUrl":87},"references\u002Frunning.md",20051,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002Freferences\u002Frunning.md",{"path":89,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":90,"sourceHtmlUrl":91},"references\u002Fsetup.md",30886,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002Freferences\u002Fsetup.md",{"path":93,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":94,"sourceHtmlUrl":95},"references\u002Fwriting-tests.md",36611,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftester-army\u002Fe2e\u002Fskills\u002Fe2e\u002Freferences\u002Fwriting-tests.md",[97,107,115,123,131,139],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":101,"licenseFileUrl":102,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":106},"01M7DV332YQMGD8WTRJ9DZQXTB","worktable-review-with-annotations","Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.","MIT","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":101,"licenseFileUrl":111,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":106},"01M7DV2YGHCZVBEB1AHYB6CGHE","worktable-find-and-synthesize-context","Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FSKILL.md",[],[],{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":101,"licenseFileUrl":119,"sourceHtmlUrl":120,"categories":121,"keywords":122,"publishedAt":106},"01M7DV2TQN1H11ME2SAF9MXN7V","worktable-create-or-update-docs","Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FSKILL.md",[],[],{"skillNo":124,"name":125,"nameZh":6,"description":126,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":101,"licenseFileUrl":127,"sourceHtmlUrl":128,"categories":129,"keywords":130,"publishedAt":106},"01M7DV2NE452NEA8TXATKSNAJH","worktable-create-or-manage-records","Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FSKILL.md",[],[],{"skillNo":132,"name":133,"nameZh":6,"description":134,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":101,"licenseFileUrl":135,"sourceHtmlUrl":136,"categories":137,"keywords":138,"publishedAt":106},"01M7DV2HENRBR2KSCVM064WH0Y","worktable-create-or-manage-interactive-html","Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FSKILL.md",[],[],{"skillNo":140,"name":141,"nameZh":6,"description":142,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":101,"licenseFileUrl":143,"sourceHtmlUrl":144,"categories":145,"keywords":146,"publishedAt":106},"01M7DV2CADR5R7RQFFYPNAGR9Y","worktable-collaborate-threads","Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FSKILL.md",[],[]]