Server data from the Official MCP Registry
Free-first SEO for agents: GSC, Yandex Webmaster/Metrika, Bing, CrUX, PSI, crawl.
About
Free-first SEO for agents: GSC, Yandex Webmaster/Metrika, Bing, CrUX, PSI, crawl.
Security Report
The lv24-seo server is a free-first SEO tool whose network access to Google Search Console, Yandex, Bing, CrUX and PSI matches its stated purpose. Provider tokens are read from the OS keychain or environment variables, and writes are gated behind an explicit allow_write flag, a plan hash and a confirm token. The main items to review are the subprocess use in the scheduler and host-config code, and a fallback write secret that may be empty when its environment variable is unset; the shown code is truncated, so those paths were not fully traced. Supply chain analysis found 10 known vulnerabilities in dependencies (0 critical, 8 high severity). Package verification found 1 issue.
4 files analyzed Β· 14 issues found
Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.
Permissions Required
This plugin requests these system permissions. Most are normal for its category.
How to Install
Add this to your MCP configuration file:
{
"mcpServers": {
"lv24-seo": {
"command": "uvx",
"args": [
"lv24-seo==0.8.0"
]
}
}
}Some servers take additional arguments. See the repository for details.
Documentation
View on GitHubFrom the project's GitHub README.

lv24-seo-optimizer
Free-first SEO control plane for humans and AI agents.
Your site's SEO, without a dashboard. Pull Search Console, Yandex Webmaster and Metrika, Bing, CrUX and your own crawl into one history β then get a weekly diff by email, an MCP server and a CLI that agents can actually use. No paid platform, no SERP scraping, no vendor account.
| π Start | π§ Use | π Operate |
|---|---|---|
| Install Β· See it work Β· Desktop | CLI Β· MCP tools Β· Skills | Scheduling Β· Data limits Β· Security |
One tool that pulls Google Search Console, Yandex Webmaster, Yandex Metrika, Bing Webmaster, Chrome UX Report and PageSpeed Insights into a single normalized model, crawls your site for technical SEO regressions, and turns all of it into scheduled email reports, MCP tools, CLI commands and agent skills.
No paid SEO platform required. Everything in the default path has a free tier.
βββββββββββββ data sources (free) βββββββββββββ βββββ surfaces βββββ
β GSC Β· Yandex Webmaster Β· Metrika Β· Bing Β· β ββββΆ β CLI (--json) β
β CrUX Β· PSI Β· your own crawl + nginx logs β β MCP server β
ββββββββββββββββββββββββββββββββββββββββββββββββ β Skills / AGENTS β
β Email / Telegram β
β SQLite history β
ββββββββββββββββββββ
Why another SEO tool
- Free-first fusion. Existing MCP servers wrap one vendor each and hand raw vendor shapes to the model. We join GSC + Bing + Yandex + Metrika per property, and we say out loud what the APIs cannot give you (Yandex has no queryΓpage join, GSC caps at 50k rows/day, data lags 2β3 days). Bounded provider responses expose coverage, freshness and limitations; unknown totals stay unknown.
- Comparable periods. The headline compares current and previous GSC windows. Yandex, Metrika, Bing, CrUX field data and PSI lab data keep separate sections. Opt-in delivery applies the configured significance thresholds and skips content already sent for the period.
- Agent-native. A small, stable MCP tool surface (no provider writes by default), a JSON envelope on every CLI command, and skills that encode audit β plan β fix β verify against a real repository.
- Zero spend. Public-repo CI, SQLite storage, SMTP/Telegram delivery, no SaaS in the default path. Paid-provider integrations are research/planned work, not dependencies or shipped adapters.
Status
Beta (v0.9.0). Working today: six
provider adapters (GSC, Yandex Webmaster, Yandex Metrika, Bing/IndexNow, CrUX, PSI), a 30-rule
crawler, SQLite history, analysis (deltas, opportunities, indexing triage, multi-engine fusion,
nginx log analysis with UA-only bot triage, CWV by template), per-provider stats, report builders
- HTML/text/Markdown renderers with an attributed change digest and a Core Web Vitals section, the
finding/task lifecycle, portable backup bundles with non-destructive and two-phase restore, a
read-only multi-site portfolio scan, SMTP/Telegram delivery with idempotency, Bing sitemap
submission, one-click IndexNow from the project sitemap, the 22-command CLI, the MCP server
(12 tools, 3 resources, 3 prompts) and a plan-by-default scheduler lifecycle. Canonical source ids,
GSC-only headline deltas, partial provider isolation and window/daily storage separation are
enforced with regression tests.
0.9.0is the release candidate for Bing sitemap submit and one-click IndexNow;0.8.0remains the latest published build on PyPI, npm and the MCP Registry (2026-11-16). Seedocs/releases/0.9.0.md,docs/releases/0.8.0.mdanddocs/releases/0.7.0.md.
Still honest gaps before v1.0: GSC sitemap submit, Bing sitemap delete and Yandex recrawl return
an explicit WRITE_DISABLED (Bing sitemap submit and IndexNow really submit), and
raw_retention_days is reserved (no raw archives are written yet). GSC supports both a plain OAuth
access token and a service-account JSON key. See
docs/plans/2026-11-10-implementation-plan.md for
the wave status.
Install
# Python (recommended)
uvx --from lv24-seo lv24-seo doctor
pipx install "lv24-seo[secrets]" # or: uv tool install "lv24-seo[secrets]"
# Node users: thin launcher over the same CLI
npx @lv24/seo doctor --json
lv24-seo init --property https://example.com # writes seo.toml + data/
lv24-seo auth set --provider gsc --stdin # token to the OS keychain (needs lv24-seo[secrets])
lv24-seo auth set --provider metrika --stdin # Metrika has its own auth provider and token
lv24-seo gsc query --start 2026-11-01 --end 2026-11-07 --dimensions query,page
lv24-seo report weekly --days 7 --out reports/
lv24-seo report diff --markdown # week-over-week diff of stored snapshots
lv24-seo stats --days 7 --json # read-only per-provider dashboard snapshot
lv24-seo crawl --json # issues + sitemapβcrawl coverage diff
lv24-seo logs analyze --file /var/log/nginx/access.log --json # bot rollups + crawl waste (read-only)
lv24-seo tasks list --json # local finding lifecycle (open β verified)
lv24-seo portfolio list --root "$HOME/sites" --json # read-only roll-up over sibling projects
mkdir -p backups
lv24-seo backup create --out "$PWD/backups" --json # portable zip: config + SQLite history
lv24-seo backup restore --file "$PWD/backups/lv24-seo-backup-<stamp>.zip" \
--to "$PWD/restored" --json # new dir; existing files untouched, no token
lv24-seo backup restore --file "$PWD/backups/lv24-seo-backup-<stamp>.zip" \
--config "$PWD/seo.toml" --json # plan; add --allow-write --confirm TOKEN to apply
lv24-seo export skills --out ./skills # packaged agent skills from the wheel
lv24-seo mcp serve # stdio MCP server
lv24-seo install --client claude|opencode|vscode|cursor
npx @lv24/seorequires uv (or an installedlv24-seobinary): the npm package is a launcher, not a second implementation.
Desktop app
The same engine ships as LV24 SEO Optimizer Desktop 0.6.0 β a Tauri v2 shell with a dashboard,
weekly report view, configuration editor/diff/validation and a recommendations-to-agent workflow. It
bundles a pinned lv24-seo sidecar (0.8.0 in the current desktop build) and adds no provider logic
of its own.
| Platform | Assets | Auto-update |
|---|---|---|
| Windows 10/11 x64 | NSIS LV24.SEO.Optimizer_0.6.0_x64-setup.exe (+ .exe.sig), LV24-SEO-Optimizer_0.6.0_x64-portable.zip | NSIS only |
| Linux x64 | LV24.SEO.Optimizer_0.6.0_amd64.AppImage (+ .AppImage.sig), LV24.SEO.Optimizer_0.6.0_amd64.deb, LV24.SEO.Optimizer-0.6.0-1.x86_64.rpm | AppImage only |
| macOS 13+ arm64/x64 | LV24.SEO.Optimizer_0.6.0_aarch64.dmg, LV24.SEO.Optimizer_0.6.0_x64.dmg, .app.tar.gz (+ .sig) | .app only |
deb, rpm and the portable ZIP are manual-only (MSI is not produced); 0.1.0 has no updater. Releases
publish latest.json and SHA256SUMS.txt.
One secret store. The desktop and the CLI use the same OS keychain entry per provider (and per
delivery secret): service lv24-seo, usernames gsc_token, yandex_webmaster_token,
yandex_metrika_token, bing_api_key, crux_api_key, psi_api_key, smtp_password,
telegram_bot_token. A token entered with lv24-seo auth set in a terminal appears in the desktop's
auth status, and a token entered in the desktop is visible to the CLI. LV24_SEO_* environment
variables are inherited by the desktop's CLI runner. The opt-in <data-dir>/secrets.json fallback
(0600 on POSIX, ACLs on Windows) is shared only when both surfaces use the same workspace/data
directory.
Desktop repo: https://github.com/LevapVeeskela/lv24-seo-optimizer-desktop Β· design notes:
docs/plans/2026-11-14-seo-desktop.md.
See it work
$ lv24-seo doctor
[OK ] python: 3.13.12
[OK ] platform: win32
[OK ] config: C:\work\seo.toml
[OK ] cwd_writable: C:\work
[FAIL] tokens: none present; run `lv24-seo auth status`
doctor checks the runtime, config, data dir, working-directory writability and token presence only
β no network calls.
$ lv24-seo report weekly --days 7 --json --out reports/
# Writes report.json and report.html; the JSON envelope carries:
# meta.sources successful providers
# meta.sources_failed isolated source failures
# meta.complete false for missing, capped or sampled inputs
# meta.limitations caveats to read before interpreting the numbers
# No configured tokens: an explicit no-data report, with complete=false.
$ lv24-seo schedule install --kind weekly --config ./seo.toml
# dry_run: true, installed: false β prints the cron block / systemd units / task XML only.
$ lv24-seo schedule install --kind weekly --config ./seo.toml --allow-write
# dry_run: false, installed: true β writes the managed cron block / user units / task.
$ lv24-seo schedule status --kind weekly --config ./seo.toml
# state: installed, matches: true (read-only)
The same data is available as MCP tools (seo_weekly_report, gsc_search_analytics,
yandex_query_stats, seo_indexing_triage, β¦) and as three skills that encode
audit β plan β fix β verify.
What it will do
| Surface | Examples |
|---|---|
| Reports | weekly search performance + delta, indexing triage, striking-distance opportunities, content decay, technical regressions, and a Core Web Vitals section fed by cwv collect (CrUX URL+device p75, PSI lab) |
| MCP tools | 12 tools across GSC, Yandex, crawl, reports, logs and gated writes β see the table below |
| CLI | 22 top-level commands (15 groups + 7 commands): version, doctor, init, properties, stats, crawl, install, auth, config, gsc, yandex, sitemap, indexnow, report, cwv, schedule, export, logs, tasks, backup, portfolio, mcp; read commands accept --strict to exit 5 on truncated data |
| Skills | seo-weekly, seo-indexing-triage, seo-technical-audit β decision trees + tool recipes + output contracts (install from the wheel: lv24-seo export skills --out ./skills) |
| Automation | cron / systemd timer / Windows Task Scheduler generators with plan-first schedule install (--allow-write to apply), read-only schedule status, exact and idempotent schedule uninstall, plus a GitHub Actions recipe |
MCP tools
Twelve tools, three resources and three prompts. Ten are read-class: seven pure read-only tools that
never change provider or local state, plus seo_crawl_site, seo_weekly_report and
yandex_indexing_state, which persist local snapshots or indexing events and therefore carry
readOnlyHint: false (they are not destructive). The two provider writes require allow_write=true,
are annotated destructiveHint: true, and use the same two-phase plan_hash / confirm_token flow
and LV24_SEO_WRITE_SECRET as the CLI; Bing sitemap submit is wired, while GSC submit, Bing delete
and Yandex recrawl return WRITE_DISABLED and nothing is sent.
| Tool | Kind | readOnlyHint | destructiveHint | What it does |
|---|---|---|---|---|
seo_properties | read | true | false | Configured providers, targets and token presence |
gsc_search_analytics | read | true | false | GSC query/page/country/device rows |
yandex_query_stats | read | true | false | Top Yandex queries for the host |
yandex_indexing_state | read + local state | false | false | Indexing history, events and problems; persists provider events when a store exists |
seo_crawl_site | read + local state | false | false | Bounded crawl; stores pages and issues (off-loop, 600 s budget) |
seo_weekly_report | read + local state | false | false | Weekly digest; stores a report snapshot |
seo_report_diff | read | true | false | Diff two stored weekly snapshots |
seo_log_analysis | read | true | false | One local nginx JSON log: bot rollups and crawl-waste share (UA-only triage) |
seo_indexing_triage | read | true | false | Group indexing problems into an action list |
seo_opportunities | read | true | false | Striking-distance and decay opportunities |
seo_recrawl_request | write | false | true | Queue Yandex recrawl URLs (two-phase gate; WRITE_DISABLED in this build) |
seo_sitemap_submit | write | false | true | Submit a sitemap to Bing or GSC (two-phase gate; Bing submit wired, GSC/delete WRITE_DISABLED) |
Resources: seo://config, seo://report/latest, seo://docs/tools. Prompts: weekly_seo_review,
indexing_triage, content_opportunities. The full inputs, outputs and limitations are in
docs/mcp.md; the generated contract is schemas/tools.json.
Supported sources
| Source | Free tier | Used for | Write operations |
|---|---|---|---|
| Google Search Console | yes (OAuth access token or service-account JSON key) | queries, pages, countries, devices; URL Inspection adapter | sitemap submit disabled |
| Yandex Webmaster | yes (OAuth) | queries, indexing history, diagnostics | recrawl disabled |
| Yandex Metrika | yes | sessions and users; filtered reports through adapter; separate metrika auth provider (yandex_metrika_token) | β |
| Bing Webmaster + IndexNow | yes | queries, crawl issues, sitemap submit, one-click IndexNow from the sitemap | sitemap submit + IndexNow submit |
| Chrome UX Report | yes (API key) | origin-level field CWV | β |
| PageSpeed Insights | yes | lab audits per URL | β |
| Your site | β | crawl, robots, sitemap, JSON-LD, hreflang, links | β |
| nginx access logs | β | logs analyze + seo_log_analysis: bot/day rollups and crawl-waste share (UA-only triage, 100 MiB cap); nothing stored | β |
Documents
- Implementation plan β waves, work packages, DoD
- Architecture β layers, data model, provider contracts
- Data sources & limits β what each API can and cannot do
- MCP tool surface β tool names, inputs, outputs, annotations
- CLI reference β commands, JSON envelope, exit codes
- Configuration β
seo.toml, env vars, secret storage - Scheduling & delivery β cron/systemd/Task Scheduler, SMTP, Telegram
- Agent skills β skill format and mirrors
- Release notes 0.9.0 β Bing sitemap submit, one-click IndexNow (RC)
- Release notes 0.8.0 β CWV writer, Yandex indicators, safe restore, schedule lifecycle
- Release notes 0.7.0 β coverage release: portfolio, restore, indexing events, CWV
- Release notes 0.6.0 β insights release: logs, digest, tasks, backup
- Release notes 0.5.3 β P0 hardening release
- Desktop app β Tauri shell, 0.6.0 assets and update matrix
- Desktop plan β shell scope and milestones
- Security β token handling, read-only defaults, write gating
- Research 2026 β API limits, existing MCP servers, tool landscape
For AI agents
Read AGENTS.md first. It is the canonical contract: repo layout, how to run things,
what is read-only, how to propose a change. Machine-readable helpers:
schemas/tools.jsonβ generated tool schemas (works without MCP)llms.txtβ doc map for LLM crawlersskills/*/SKILL.mdβ portable skills (Claude Code, OpenCode, Codex)- MCP Registry:
io.github.LevapVeeskela/lv24-seo
License
Apache-2.0. See LICENSE and TRADEMARKS.md. Data pulled from Google/Yandex/Bing belongs to the property owner β this project never resells it.
Reviews
No reviews yet
Be the first to review this server!
More Developer Tools MCP Servers
Rea
Freeby Morluto Β· Developer Tools
Reverse engineer anything from your terminal or agent with one CLI and MCP server.
Paperclip
Freeby Paperclipai Β· Developer Tools
Trending hip-hop artist momentum scores across four cultural dimensions.
Scrapling
Freeby D4Vinci Β· Developer Tools
Web scraping with stealth HTTP, real browsers, and Cloudflare bypass. CSS selectors supported.
Git
Freeby Modelcontextprotocol Β· Developer Tools
Read, search, and manipulate Git repositories programmatically
Toleno
Freeby Toleno Β· Developer Tools
Toleno Network MCP Server β Manage your Toleno mining account with Claude AI using natural language.
MCP Marketplace
Freeby mcp-marketplace Β· Developer Tools
Search and install MCP servers from inside your AI client.
