context-attachment-manager
Codex and Claude Code plugin (works in Claude Desktop): choose which past images go to the model with your next message. Unchecked images become placeholders, the conversation goes on, and requests st
Full-spectrum, evidence-first SEO plugin for Claude Code
Translation pending
PluginsPythonGitHub @ChaoticSurferOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
Full-spectrum, evidence-first SEO plugin for Claude Code: 18 skills, stdlib Python auditors, code-computed triage, beginner & expert guides, cost-aware Semrush/Ahrefs/DataForSEO use
Translation pending
Taken from the repository README
Loading the full README.
Read the full README ↗A Claude Code plugin that does SEO end to end, from a five-minute page check to a full audit, migration, or AI-search visibility programme. It is built so the free path always works and paid tools are an upgrade with a stated reason.
seo-suite router plus audit, technical, onpage, content, blog-builder, blog-review,
blog-system, keywords, schema, ai-search, local, international, ecommerce, media, links, migration, analytics,
competitors, programmatic, strategy and automation.seo-site-auditor collects data and seo-serp-researcher runs SERP batches in parallel, both on
Sonnet; seo-content-reviewer grades pages on the session's model. No agent uses Haiku.translation_check.py)
and AI-text quality heuristics with no AI-detection score (ai_text_check.py). build_audit.py audits a static
build folder offline, and causal_claims.py flags report sentences that call a change the cause of a metric move.claude_seo.py). On first need seo-suite clones
AgriciDaniel/claude-seo into its own data folder, pulls its latest main
before every use, and runs its tools: the Playwright renderer for JavaScript sites, its Search Console client
(Search Analytics, URL Inspection, sitemaps; never the Indexing API) and its drift rules between crawls. It is not
added as a second plugin.schema_hook.py). After each Edit or Write of a markup file, JSON-LD that doesn't
parse, a leftover placeholder or a missing @type goes back to Claude to fix; Google's missing fields come back as
advice. Other files are skipped at once.research_store.py). Keyword, search-results and Search Console research is kept
in one folder per site, so it is done once: every skill reads it first, re-fetches only what has gone stale, and
writes new findings back.credit_guard.py). It runs as a hook around every paid API call. It reuses cached results,
limits rows and keeps spending inside the monthly budgets you set. Every budget starts at 0, so paid calls are
off until you choose otherwise.audit_profile.py). An audit picks its tool stack and depth by fixed rules: the free
big-website stack by default at every size, the paid list for the site's size only when a DataForSEO or other-tools
budget (or Ahrefs, on big sites) is above 0, audit_profile is paid, or you ask for it, and a quick one-URL check
only on request (references/profiles.md).google-rules.md: dated, verified Google and Bing rules, with an explicit list of unverified items.tools.md: the default free stack every run uses, the labels (Main, Cross-check, Reserve) and the rules behind
every list. tools-paid.md: the paid stacks for small and big sites, each category's Reserve table, the
credit-guard limits, why each paid tool is worth it and how to spend less on it. tools-sources.md: their sources.guide-beginner.md and guide-expert.md: the same SEO pillars at two depths. The router picks one
for the user's level.autonomy.md: what Claude decides on its own, what it asks, and the autonomy levels.paid-tools-automation.md: how Semrush, Ahrefs and DataForSEO are automated, and the credit guard in detail.research-store.md: the research store's layout, freshness rules and commands.ai-content-review.md: review notes for AI-written and AI-translated content.scripts.md: every script, its inputs and its output.worklog.md: the SEO-WORKLOG.md entry format that every task appends to.In Claude Code:
/plugin install seo-suite --marketplace ChaoticSurfer/seo-suite
Or from a terminal:
claude plugin marketplace add ChaoticSurfer/seo-suite
claude plugin install seo-suite@seo-suite
You need Python 3.9 or newer. The scripts use only the standard library, so there is nothing else to install by hand. Rendering, the Search Console API and technical drift use claude-seo, which seo-suite installs itself the first time a run needs it (a venv and Chromium, about 150 MB). That needs git and Python 3.11–3.14.
Everything works on free data. To add a paid source, install its connector:
/plugin install seo-suite-semrush@seo-suite
/plugin install seo-suite-ahrefs@seo-suite
/plugin install seo-suite-dataforseo@seo-suite
Then sign in: /mcp for Semrush and Ahrefs, and /plugin configure seo-suite-dataforseo@seo-suite for
DataForSEO's API login. Set the monthly budgets with /plugin configure seo-suite@seo-suite. Until you do, the
credit guard blocks every paid call.
INTEGRATIONS.md lists every tool the plugin works with, free and paid, how each one connects, and what it costs.
A run goes end to end without stopping to ask. Claude makes the SEO decisions itself and asks only for access it
lacks (a sign-in, an export) and for real business decisions, all together at the end. What it may change depends
on the autonomy level you set with /plugin configure seo-suite@seo-suite:
| Level | Runs without asking |
|---|---|
| 1. Analyse (default) | Fetches, crawls, scripts, free APIs, paid calls within budget, reports, the work log |
| 2. Fix in code | Level 1, plus edits, commits and pull requests in the site's repo |
| 3. Ship | Level 2, plus merging pull requests that pass CI, and submitting sitemaps or IndexNow for changed URLs |
| 4. Operate | Level 3, plus publishing to a CMS and Search Console sitemap submission through an authorised connector |
Some actions always need your OK, at every level: messages to real people, disavow files, deleting pages that earn traffic, DNS and CDN changes, Business Profile and Merchant Center edits, and spending above budget. The run prepares each one exactly and lists it at the end; Claude carries out the items you approve. Buying reviews or ranking links, and anything a vendor's terms forbid, are never done. The full list is in references/autonomy.md.
Ask in plain words: "audit example.com", "why did traffic drop in September", "add Product schema",
"get us cited in ChatGPT", "teach me SEO". You can also call a skill directly: /seo-suite:seo-audit.
Scripts run on their own too. <plugin-dir> is your clone, ~/.claude/skills/seo-suite for a local copy, or
~/.claude/plugins/cache/seo-suite/seo-suite/<version>/ after a marketplace install:
python3 <plugin-dir>/scripts/page_audit.py https://example.com/
python3 -m unittest discover -s <plugin-dir>/tests
test_docs.py fails if a skill references a script or flag that doesn't exist, or a section of google-rules.md
has gone unchecked for about six months. test_scripts.py fails if a script can emit a finding id that has no triage
row. GitHub Actions runs the suite on every push (.github/workflows/tests.yml) and prints the test count in the job
summary.
claude plugin disable seo-suite@seo-suite, or
claude plugin uninstall seo-suite@seo-suite.~/.claude/skills/seo-suite loads automatically as seo-suite@skills-dir. Delete the folder
to remove it.MIT