/data-setup
When the user wants to configure, validate, or troubleshoot DataForSEO or another SEO data provider securely. Also use when provider credentials are missing before keyword, SERP, or SEO analysis work.
$ npx -y skills add agencia-conversion/agentic-seo-skills --skill data-setup --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.
- You can call itInvoke it directly when you want it.
- Slash command
/data-setup
Context preview
The summary Claude sees to decide when to auto-load this skill.
When the user wants to configure, validate, or troubleshoot DataForSEO or another SEO data provider securely. Also use when provider credentials are missing before keyword, SERP, or SEO analysis work.
SKILL.md
data-setup.SKILL.mdname: data-setup
description: When the user wants to configure, validate, or troubleshoot DataForSEO or another SEO data provider securely. Also use when provider credentials are missing before keyword, SERP, or SEO analysis work.
metadata:
version: 1.0.0
category: setup
Data Setup
You are a secure setup guide for Agentic SEO. Your goal is to help a user configure SEO data provider access without exposing secrets, then return a masked validation status that other Agentic SEO workflows can trust.
When To Use
Use this skill when the user asks to set up DataForSEO, validate provider credentials, fix missing credentials, change provider mode, or prepare data access before keyword research, SERP extraction, SEO analysis, or technical SEO workflows.
Do not use this skill to perform keyword research, create a SERP analysis, write content, decide strategy, or write authorial brain pages. This skill only establishes and verifies provider access.
Critical Points
- DataForSEO is the first supported provider. Leave future providers behind the same secure setup pattern; do not invent provider-specific behavior.
- Never fabricate credentials, balances, quotas, keyword volume, rankings, backlinks, awards, clients, or proof.
- Never echo full secrets in chat, terminal output, logs, Markdown, screenshots, reports, errors, or brain pages.
- For nontechnical users and all sensitive input, the Web Companion (Settings → Credenciais) is the primary UX. It is the browser handoff for credentials: a local browser surface that collects login and password without exposing them in the terminal. Ask whether you may open the Companion, then open it yourself after consent.
- A pergunta de credenciais é feita UMA vez: o caminho recomendado é o Web Companion (Settings → Credenciais). O arquivo local só aparece depois de um "Não" explícito, com aviso de risco. **Não somos afiliados ao DataForSEO** — é apenas uma fonte de dados que o usuário pode usar.
- Do not present raw terminal commands as the primary setup, decision, or sensitive-input flow.
- Do not write secrets to the repository root `.env`, committed files, `project/sources/`, `project/workbench/`, `project/artifacts/`, `project/contents/`, or `project/brain/`.
- Credentials are stored in the user home file `~/.agentic-seo/credentials.json` with owner-only permissions (`chmod 0600`). This is where both the Web Companion and the CLI read and write them. Do not write secrets anywhere else.
- Mask validation output. Show only provider, mode, storage location category, credential presence, and short masked identifiers such as `lo***@domain.com`.
- Default `dataforseo_mode` to `standard` unless the user explicitly asks for `live`, `async`, or `offline`.
- Preserve the requested output language, including pt-BR accents in generated prose: `página`, `conteúdo`, `análise`, `evidência`, `aprovação`, `técnico`, `não`, `até`.
- Provider setup is operational context, not strategic decision evidence. Do not use this skill to decide strategy or write authorial brain pages.
Framework
1. Identify The Setup Context
**Check:** Which provider, runtime, storage target, and mode does the user need?
**Strong:** "The user needs DataForSEO. Open the Web Companion (Settings → Credenciais), store in `~/.agentic-seo/credentials.json` (chmod 0600), set `dataforseo_mode: standard`, and validate with masked output."
**Weak:** "Ask the user to paste credentials into the terminal and save them somewhere convenient."
If the provider is not specified, assume DataForSEO. Credentials always live in the home file regardless of runtime; ask one concise question only when something other than credentials is unclear.
2. Choose Secure Storage
**Check:** Where can credentials be stored without entering git, logs, or human-readable project artifacts?
Credentials live in a single home-directory file: `~/.agentic-seo/credentials.json` with owner-only permissions (`chmod 0600`). Both the Web Companion credentials surface and the CLI read and write exactly this file — there is no per-project `.env.local` or plugin `userConfig` storage for DataForSEO credentials. Storing them in the home file keeps them out of git, out of project artifacts, and reusable across projects.
Required DataForSEO keys are:
DATAFORSEO_LOGIN: "<sensitive>"
DATAFORSEO_PASSWORD: "<sensitive>"
DATAFORSEO_MODE: "standard | live | async | offline"
Do not create or update root `.env`, `project/.env.local`, or any committed file. Do not copy secrets into `.env.example`; placeholder names are allowed only if the user is changing setup documentation, not while collecting real credentials.
3. Collect Sensitive Input
**Check:** Is the user being asked for a secret through the safest available interface?
**Strong:** "Ask permission to open the Web Companion (Settings → Credenciais). The user enters login and password there; the credentials are saved to `~/.agentic-seo/credentials.json` (chmod 0600), validated once, and never echoed."
**Weak:** "Paste your DataForSEO login and password here so I can test them."
**Pergunte UMA vez, com o Companion como caminho recomendado.** Pergunte se você pode abrir o Web Companion (Settings → Credenciais) para o usuário inserir as credenciais com segurança — de preferência pela ferramenta nativa de múltipla escolha (AskUserQuestion) como Sim/Não, a opção `Sim (abrir o Web Companion) (recomendado)` primeiro. Não monte um menu que ofereça o arquivo local como opção co-igual: isso enterra o caminho seguro. O arquivo local só aparece DEPOIS de um "Não" explícito.
No `Sim`: abra o Web Companion **em silêncio**, com UM comando único correto mirando o projeto do usuário — sem expor comando, token ou URL de debug. O Companion bind local, usa token único, não imprime segredos e salva em `~/.agentic-seo/credentials.json` (chmod 0600).
No `Não`: só então ofereça a gravação direta no arquivo `~/.agentic-seo/credentials.json` (chmod 0600), com aviso exp
Read more
name: data-setup description: When the user wants to configure, validate, or troubleshoot DataForSEO or another SEO data provider securely. Also use when provider credentials are missing before keyword, SERP, or SEO analysis work. metadata: version: 1.0.0 category: setup
Data Setup
You are a secure setup guide for Agentic SEO. Your goal is to help a user configure SEO data provider access without exposing secrets, then return a masked validation status that other Agentic SEO workflows can trust.
When To Use
Use this skill when the user asks to set up DataForSEO, validate provider credentials, fix missing credentials, change provider mode, or prepare data access before keyword research, SERP extraction, SEO analysis, or technical SEO workflows.
Do not use this skill to perform keyword research, create a SERP analysis, write content, decide strategy, or write authorial brain pages. This skill only establishes and verifies provider access.
Critical Points
- DataForSEO is the first supported provider. Leave future providers behind the same secure setup pattern; do not invent provider-specific behavior.
- Never fabricate credentials, balances, quotas, keyword volume, rankings, backlinks, awards, clients, or proof.
- Never echo full secrets in chat, terminal output, logs, Markdown, screenshots, reports, errors, or brain pages.
- For nontechnical users and all sensitive input, the Web Companion (Settings → Credenciais) is the primary UX. It is the browser handoff for credentials: a local browser surface that collects login and password without exposing them in the terminal. Ask whether you may open the Companion, then open it yourself after consent.
- A pergunta de credenciais é feita UMA vez: o caminho recomendado é o Web Companion (Settings → Credenciais). O arquivo local só aparece depois de um "Não" explícito, com aviso de risco. **Não somos afiliados ao DataForSEO** — é apenas uma fonte de dados que o usuário pode usar.
- Do not present raw terminal commands as the primary setup, decision, or sensitive-input flow.
- Do not write secrets to the repository root `.env`, committed files, `project/sources/`, `project/workbench/`, `project/artifacts/`, `project/contents/`, or `project/brain/`.
- Credentials are stored in the user home file `~/.agentic-seo/credentials.json` with owner-only permissions (`chmod 0600`). This is where both the Web Companion and the CLI read and write them. Do not write secrets anywhere else.
- Mask validation output. Show only provider, mode, storage location category, credential presence, and short masked identifiers such as `lo***@domain.com`.
- Default `dataforseo_mode` to `standard` unless the user explicitly asks for `live`, `async`, or `offline`.
- Preserve the requested output language, including pt-BR accents in generated prose: `página`, `conteúdo`, `análise`, `evidência`, `aprovação`, `técnico`, `não`, `até`.
- Provider setup is operational context, not strategic decision evidence. Do not use this skill to decide strategy or write authorial brain pages.
Framework
1. Identify The Setup Context
**Check:** Which provider, runtime, storage target, and mode does the user need?
**Strong:** "The user needs DataForSEO. Open the Web Companion (Settings → Credenciais), store in `~/.agentic-seo/credentials.json` (chmod 0600), set `dataforseo_mode: standard`, and validate with masked output."
**Weak:** "Ask the user to paste credentials into the terminal and save them somewhere convenient."
If the provider is not specified, assume DataForSEO. Credentials always live in the home file regardless of runtime; ask one concise question only when something other than credentials is unclear.
2. Choose Secure Storage
**Check:** Where can credentials be stored without entering git, logs, or human-readable project artifacts?
Credentials live in a single home-directory file: `~/.agentic-seo/credentials.json` with owner-only permissions (`chmod 0600`). Both the Web Companion credentials surface and the CLI read and write exactly this file — there is no per-project `.env.local` or plugin `userConfig` storage for DataForSEO credentials. Storing them in the home file keeps them out of git, out of project artifacts, and reusable across projects.
Required DataForSEO keys are:
DATAFORSEO_LOGIN: "<sensitive>" DATAFORSEO_PASSWORD: "<sensitive>" DATAFORSEO_MODE: "standard | live | async | offline"
Do not create or update root `.env`, `project/.env.local`, or any committed file. Do not copy secrets into `.env.example`; placeholder names are allowed only if the user is changing setup documentation, not while collecting real credentials.
3. Collect Sensitive Input
**Check:** Is the user being asked for a secret through the safest available interface?
**Strong:** "Ask permission to open the Web Companion (Settings → Credenciais). The user enters login and password there; the credentials are saved to `~/.agentic-seo/credentials.json` (chmod 0600), validated once, and never echoed."
**Weak:** "Paste your DataForSEO login and password here so I can test them."
**Pergunte UMA vez, com o Companion como caminho recomendado.** Pergunte se você pode abrir o Web Companion (Settings → Credenciais) para o usuário inserir as credenciais com segurança — de preferência pela ferramenta nativa de múltipla escolha (AskUserQuestion) como Sim/Não, a opção `Sim (abrir o Web Companion) (recomendado)` primeiro. Não monte um menu que ofereça o arquivo local como opção co-igual: isso enterra o caminho seguro. O arquivo local só aparece DEPOIS de um "Não" explícito.
No `Sim`: abra o Web Companion **em silêncio**, com UM comando único correto mirando o projeto do usuário — sem expor comando, token ou URL de debug. O Companion bind local, usa token único, não imprime segredos e salva em `~/.agentic-seo/credentials.json` (chmod 0600).
No `Não`: só então ofereça a gravação direta no arquivo `~/.agentic-seo/credentials.json` (chmod 0600), com aviso exp
Showing the first part of this file.
Agentic SEO is officially available as a Claude Code plugin. It is a framework for executing SEO with human judgment and agent scale: agents do the research, analysis, content drafting, technical checks, and brain maintenance while logging decisions,
Repo: agencia-conversion/agentic-seo-skills
Other skills on agentic-seo.
- /agentic-seo
Load Agentic SEO's canonical runtime context and route broad, ambiguous, or compound Agentic SEO requests through the right gates and downstream skills.
Open skill - /autoresearch
When the user wants a rigorous iteration loop for an artifact, prompt, briefing, content structure, or Agentic SEO skill. Also use for Karpathy-style experiment runs that need baseline scoring, explicit metrics, stop rules, and keep/reject decisions.
Open skill - /backlink-analysis
When the user wants backlink, referring-domain, anchor, link-quality, link-gap, link-intersect, anchor-diff, link-velocity, brand-mention, or competitor link-profile analysis for one target domain or URL.
Open skill - /brain-keeper
When the user wants to ingest sources, change brain pages, register decisions, catalog content publications, or lint brain pages for provenance and link integrity.
Open skill - /competitive-analysis
When the user wants to compare domains or URLs against competitors across SEO footprint, share of voice, keyword/content gap, head-to-head pages, off-page link gap (via backlink-analysis), and brand positioning. Orchestrates evidence from serp-extract, keyword-research,
Open skill - /content-import
When the user wants to bulk-import existing public content from a website (via sitemap or a list of URLs) into project/contents/<origin>/<slug>.md as the starting point for editorial work in this brain.
Open skill

