[NoQA] Add agent-device glue-code skill for mobile testing #87662
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,43 @@ | ||
| --- | ||
| name: agent-device | ||
| description: Drive iOS and Android devices for the Expensify App - testing, debugging, performance profiling, bug reproduction, and feature verification. Use when the developer needs to interact with the mobile app on a device. | ||
| --- | ||
|
|
||
| # agent-device | ||
|
|
||
| ## Pre-flight Check | ||
|
|
||
| Verify the `agent-device` CLI is installed and its skills are accessible: | ||
|
|
||
| ```bash | ||
| agent-device --version | ||
| ``` | ||
|
|
||
| If missing, **STOP** and instruct the user to install it: | ||
|
|
||
| ```bash | ||
| npm install -g agent-device | ||
| ``` | ||
|
|
||
| The `agent-device` CLI ships with built-in skills under `skills/` in the installed package. These contain the canonical reference for device automation - bootstrap, exploration, verification, debugging, and more. Use `agent-device --help` to discover available commands and skill names. Read the skill files directly from the installed package path when you need detailed guidance: | ||
|
Contributor
There was a problem hiding this comment. Is this not too vague? It's not specifically told to do this at launch so I'm not sure how well it will descide to go and ready them for guidance. |
||
|
|
||
| ```bash | ||
| # Find the package location | ||
|
Contributor
There was a problem hiding this comment. I wish we could run commands automatically instead of asking the agent to do something... Locally this fails with permission though, so not sure how we could do it. Can we update the settings and allow this commands and inject the context without a tool call? |
||
| npm root -g | ||
| # Then read: <global_root>/agent-device/skills/agent-device/SKILL.md | ||
| ``` | ||
|
|
||
| > **Do not proceed without `agent-device` installed.** All device interaction depends on it. | ||
|
|
||
| ## How This Skill Works | ||
|
|
||
| This skill enhances the local development and testing workflow. The developer may provide a full test plan upfront, give step-by-step instructions interactively, or just point at a screen and ask questions. Follow their lead. | ||
|
Comment on lines
+32
to
+34
|
||
|
|
||
| ### Principles | ||
|
|
||
| - **Fail fast.** If something deviates from expectations - wrong screen, unexpected error, missing element - stop, explain what happened and suggest how to get past it. Do not silently work around problems. | ||
| - **Deviations are signal.** If the app behaves differently than expected, report the delta clearly so the developer can decide what needs fixing. | ||
|
|
||
| ## Footnote | ||
|
|
||
| > **Early development.** This integration between agent-device and the Expensify App is still in early stages. If you run into issues, please reach out via Expensify Slack or file an issue at https://github.com/callstackincubator/agent-device/issues. | ||
|
Comment on lines
+41
to
+43
Contributor
There was a problem hiding this comment. Is this more for engineers than it is to the agent?
Contributor
There was a problem hiding this comment. The idea was to start with engineers being able to kick off testing via their local agent
Contributor
There was a problem hiding this comment. Oh I mean, the footnote doesn't seem to help the agent |
||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'd like to remove
testing, debugging, performance profilingfor now. I worry Melvin will start triggering this on workflows before we have good instructions for it to follow.