ariawalk 0.1.1 → 0.3.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +27 -1
- package/dist/ariawalk.js +106 -23
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -10,9 +10,35 @@ npx ariawalk run <testId> --base-url http://localhost:3000
|
|
|
10
10
|
|
|
11
11
|
`ariawalk login` shows a code and opens ariawalk.com/device, where you confirm it while signed in. That works over SSH too: open the link on any device. The CLI then gets an API key named "CLI · <your machine>" for the organization you pick (`--org` when you're in several), which you can revoke under **API and MCP** or with `ariawalk logout`. `ariawalk whoami` shows who you're logged in as. In CI, set `ARIAWALK_API_KEY` instead.
|
|
12
12
|
|
|
13
|
+
## Commands
|
|
14
|
+
|
|
15
|
+
```sh
|
|
16
|
+
npx ariawalk tests # your tests with their ids (--json for scripts and agents)
|
|
17
|
+
npx ariawalk run # pick tests from a checklist, then run them one after another
|
|
18
|
+
npx ariawalk run <testId>… --base-url http://localhost:3000
|
|
19
|
+
npx ariawalk report <runId|url> # a run's full report as Markdown
|
|
20
|
+
npx ariawalk skills install # teach coding agents (Claude Code…) to use the CLI
|
|
21
|
+
```
|
|
22
|
+
|
|
23
|
+
The checklist is line-based (type `2 4-6` to toggle, `a` for all, Enter to run), so it works with screen readers and over SSH.
|
|
24
|
+
|
|
25
|
+
`ariawalk skills install` writes `.claude/skills/ariawalk/SKILL.md` in the current project (`--global` for every project): how to find tests, run them against the dev server, read the report and fix what it finds. Use it instead of the MCP server if your agent can run shell commands.
|
|
26
|
+
|
|
13
27
|
- `--base-url` moves the test's start URL to another origin. Localhost and private hosts need no domain verification; public hosts must be verified in Ariawalk.
|
|
14
28
|
- Secret test data values are never sent to the CLI. Pass them with `--data password=…` or `ARIAWALK_DATA_PASSWORD`.
|
|
15
29
|
- Runs from your machine land in your own "local" environment, and runs in CI (GitHub Actions, GitLab, Buildkite) land in `ci:<branch>`. They never become the baseline for production runs.
|
|
16
|
-
- Exit codes: `0`
|
|
30
|
+
- Exit codes: `0` passed, `1` an accessibility check failed, `2` a run couldn't finish (setup, quota, key, network).
|
|
31
|
+
|
|
32
|
+
## CI
|
|
33
|
+
|
|
34
|
+
```sh
|
|
35
|
+
npx ariawalk install --with-deps
|
|
36
|
+
npx ariawalk run --app "My shop" --base-url "$PREVIEW_URL" --concurrency 3 \
|
|
37
|
+
--fail-on blocked,regression --min-score 80 --junit ariawalk.xml
|
|
38
|
+
```
|
|
39
|
+
|
|
40
|
+
- `--fail-on`: `blocked` (default), `regression` (worse than the last run on the branch, else the default branch), `new-serious` (new serious or blocker issues), or `none`.
|
|
41
|
+
- `--junit`, `--markdown` and `--json-file` write results; on GitHub Actions a summary is added to the step summary automatically.
|
|
42
|
+
- Set `ARIAWALK_API_KEY` (a key with "Read and run" access) instead of logging in. CI metadata from GitHub Actions, GitLab and Buildkite puts runs in `ci:<branch>`.
|
|
17
43
|
|
|
18
44
|
Every run counts toward your monthly runs. Screenshots stay on your machine; step transcripts, findings and page snapshots are uploaded for the report.
|