@siteoshq/cli 1.10.0 → 1.12.0

package/README.md

@@ -1,7 +1,7 @@
 # SiteOS CLI
 
 `@siteoshq/cli` exposes one `siteos` binary for Auth, common Projects, Pulse, Cookie, Forms,
-Search, Trace, SEO and Integrations. This source prepares version 1.10.0; publishing is a separate release.
+Search, Trace, SEO/GEO and Integrations. This source is version 1.11.0; source changes require a separate release to reach npm.
 Node.js 22 or newer is required.
 
 ## Install and authenticate
@@ -96,6 +96,12 @@ the bound Pulse resource and current URL; local tests use this URL unless explic
 `siteos.config.json` remains the tracked Playwright check configuration. Deployment and enabling
 monitoring are explicit operations. `deploy --dry-run` builds locally without uploading.
 
+In a monorepo, run Pulse commands from the directory containing that website's
+`siteos.config.json`. The current source resolves deployment configuration, specs and archive
+output from that directory; private Project selection remains scoped to the Git repository.
+This fixes published CLI 1.10.0, which incorrectly bundled the repository root on deployment.
+Use a locally built CLI containing the fix until its separate npm release is available.
+
 CLI 1.10.0 adds an optional `environmentVariables` array to each Check, for
 example `"environmentVariables": ["TEST_EMAIL", "TEST_INBOX_TOKEN"]`. Declare names only, and set
 values in Pulse Settings for the selected Environment. The runner exposes declared values through
@@ -161,11 +167,17 @@ snippet is unavailable until runtime publication succeeds. Integrations connecti
 Organization; provider authorization opens the shared browser flow, and notification destinations
 are connected explicitly by each service.
 
-### SEO (prepared for release)
+### SEO/GEO
+
+The application calls this service **SEO/GEO**; its command and Project service key remain `seo`.
+GEO means generative engine optimization, including technical AI-search accessibility and
+AI mentions/citations. Check `siteos seo --help` against the installed CLI and server version.
 
-Check `siteos seo --help` before use; install a matching release only after it is published. An
-SEO-capable server returns SEO in its Project service catalog, which older strict CLI parsers may
-reject. Coordinate the CLI update before that server rollout, then publish the matching plugin.
+Site Audit checks public HTML, canonical, robots.txt, sitemaps, snippets and basic structured data.
+Its crawler-policy checks do not prove access from actual provider IPs. Performance adds
+selected-page Lighthouse and response/rendered HTML comparisons. Brand lookup and Prompt checks
+provide separate AI visibility evidence. No autonomous agent journey or target API/MCP audit is
+performed by these commands, and no result guarantees AI inclusion or a universal AI rank.
 
 ```sh
 siteos project connect seo --json
@@ -186,7 +198,40 @@ observed URL. Read the returned audit ID until a terminal state; a queued respon
 `issue ignore` and `issue restore` require `--audit`, `--url`, `--rule`, `--reason` and the current
 `--revision` (0 before the first decision). They use a distinct issue-write scope. Read back the
 decision after success; a revision conflict requires refreshing it. Ignore never means resolved.
-All SEO commands resolve the selected common Project and explicit environment binding. Current
+All SEO/GEO commands resolve the selected common Project and explicit environment binding.
+Research and GSC Insights use these commands with a matching server release:
+
+```sh
+siteos seo research summary --json
+siteos seo research history --kind keywords --json
+siteos seo research show <run-id> --json
+siteos seo research plan --input ./research-request.json --json
+siteos seo research run --input ./research-request.json --idempotency-key <retry-key> --json
+siteos seo research wait <run-id> --json
+siteos seo research export <run-id> --format json --output ./research-report.json --json
+siteos seo gsc status --json
+siteos seo gsc report --dataset queries --filter all --json
+siteos seo gsc export --dataset queries --format csv --output ./gsc-queries.csv --json
+```
+
+Run `siteos seo --help` for request JSON, limits, saved inputs, cancellation and GSC settings.
+Research supports keywords, Domain Overview (`domain`), rankings, backlinks, brand and AI visibility without GSC.
+Domain Overview collects domain totals, up to 100 ranking keywords, pages and competitors for a market.
+`--kind domain` includes legacy `competitors` history; existing `competitors` inputs keep their original two-part plan.
+New request example: `{"kind":"domain","target":"example.com","country":"US","language":"en"}`.
+Reading/planning/saving spends no research credits. Launch uses Organization credits and the same
+server admission as the interface. Keep the retry key after an uncertain response; no paid retries
+are automatic. Wait exit 3 means pending; exit 4 failed/cancelled. Partial evidence remains readable.
+Exports retain scope/coverage and refuse overwrite. Research CSV includes JSON cells for request,
+dataset metadata and rows; JSON is preferable for programmatic analysis.
+
+GSC uses saved comparisons and an explicitly connected property. Reports page by 25; exports include
+all matching normalized rows, up to the 20,000-row current/previous union bound. `gsc sync` queues the
+Google worker without research credits. Google OAuth remains interactive; `gsc bind`/`disconnect`
+manage only this environment with revision checks. Separate grants control research read/run/write
+and GSC read/sync/write. Audit authority alone never enables paid research. These source capabilities
+require coordinated CLI/server/plugin publication; they are not a claim about the current npm version.
+
 Auth policy permits CLI service-grant issuance to Organization owners/admins. JSON uses contract
 version 1 and does not include credentials. The focused plugin skill is `siteos-seo`.