---
title: "DataForSEO AI agent connections and job ideas"
canonical: "https://agentuse.io/ai-agents/tools/dataforseo"
last_tested: "September 2026"
---

# DataForSEO AI agent connections and job ideas

Build a DataForSEO AI agent for recurring SEO operations: collect paid crawl and ranking evidence within explicit ceilings, compare stable scopes, and prepare the changes that need human review.

Canonical HTML guide: https://agentuse.io/ai-agents/tools/dataforseo

## How to connect

MCP, CLI, API, and SDK options provide capabilities. A skill provides operating guidance only; it does not authenticate to or grant access to the service.

### DataForSEO Universal MCP Tool (MCP)

- Source: Official
- Documentation: https://dataforseo.com/help-center/dataforseo-new-mcp-server-and-cli-connection-a-quickstart-guide
- Best for: Agents that need current DataForSEO documentation and authenticated API requests
- Capability: Four tools expose the documentation index, documentation sections, documentation search, and authenticated API requests across available DataForSEO endpoints.
- Authentication: OAuth bearer authentication for hosted HTTP MCP, or DataForSEO API login and password supplied through the runtime environment for stdio
- Recommendation: Preferred starting point
- Setup reference: npx dataforseo-mcp-server
- Operating note: The current universal tool replaces the legacy catalog of more than 80 endpoint-specific MCP tools. Validate endpoint paths, request limits, and reported cost before scheduling paid calls.

### Official DataForSEO agent skill (Skill)

- Source: Official
- Documentation: https://github.com/dataforseo/mcp-server-typescript/blob/v3-mcp/SKILL.md
- Best for: CLI coding agents that should call the DataForSEO universal tool through approved shell commands
- Capability: Documents the CLI commands for reading the API documentation index, searching docs, and making authenticated requests.
- Authentication: DataForSEO API login and password supplied as runtime environment variables
- Setup reference: Install the official SKILL.md, then allowlist the exact DataForSEO CLI command shapes required by the job.
- Operating note: A skill supplies instructions, not credentials or service access. The runtime still needs the CLI package, a narrow command allowlist, and secret configuration.

### DataForSEO API v3 (API)

- Source: Official
- Documentation: https://docs.dataforseo.com/v3/auth/
- Best for: Project-owned wrappers that need exact endpoint and cost controls
- Capability: Direct HTTP access to DataForSEO API v3 products using endpoint-specific request and response contracts.
- Authentication: HTTP Basic authentication with the API login and password from the DataForSEO account
- Operating note: Keep credentials outside agent files. Allowlist exact endpoint paths and enforce per-run request, result, retry, and spend ceilings in the wrapper.

### On-Page API (API)

- Source: Official
- Documentation: https://docs.dataforseo.com/v3/on_page-overview/
- Best for: Bounded technical SEO crawls and page-level issue evidence
- Capability: Creates crawl tasks and retrieves on-page summaries, pages, resources, and issue data for an approved target.
- Authentication: DataForSEO API v3 credentials
- Operating note: Crawl size and paid operations must be bounded before a recurring agent submits a task.

### Ranked Keywords API (API)

- Source: Official
- Documentation: https://docs.dataforseo.com/v3/dataforseo_labs/google/ranked_keywords/live/
- Best for: Current domain keyword rankings for a declared search market
- Capability: Returns ranked-keyword data for a target with location, language, filters, ordering, and result limits.
- Authentication: DataForSEO API v3 credentials
- Operating note: Store the exact target, location, language, filters, limit, timestamp, and reported cost with every snapshot.

## Agent job ideas

### Weekly SEO change review

Collect a bounded crawl and ranked-keyword snapshot, compare it with the prior complete run, and reconcile observed movement with the operator change log.

- Services: [DataForSEO](https://agentuse.io/ai-agents/tools/dataforseo.md) + Change log + Evidence journal
- Expected output: SEO evidence journal
- Full tutorial: https://agentuse.io/ai-agents/workflows/track-seo-changes-with-an-ai-agent

### Technical crawl exception queue

Run one capped crawl and group new or materially changed page issues by template, status, and review owner.

- Services: [DataForSEO](https://agentuse.io/ai-agents/tools/dataforseo.md) + [Slack](https://agentuse.io/ai-agents/tools/slack.md)
- Expected output: Crawl exception artifact

### Ranking movement brief

Compare the same target and market across complete snapshots, preserving additions, removals, and unknown collection gaps.

- Services: [DataForSEO](https://agentuse.io/ai-agents/tools/dataforseo.md) + [Slack](https://agentuse.io/ai-agents/tools/slack.md)
- Expected output: Ranking movement brief

### Competitor keyword watch

Collect the same approved keyword view for a fixed competitor set and flag material coverage changes without guessing strategy.

- Services: [DataForSEO](https://agentuse.io/ai-agents/tools/dataforseo.md) + [Notion](https://agentuse.io/ai-agents/tools/notion.md)
- Expected output: Competitor change brief

### Content decay review

Join ranking movement with an approved content inventory and surface pages whose decline merits editorial investigation.

- Services: [DataForSEO](https://agentuse.io/ai-agents/tools/dataforseo.md) + [Google Drive](https://agentuse.io/ai-agents/tools/google-drive.md) + [Notion](https://agentuse.io/ai-agents/tools/notion.md)
- Expected output: Content review queue

### SEO migration evidence pack

Capture pre-launch and post-launch crawl and ranking evidence under identical scopes for a release review.

- Services: [DataForSEO](https://agentuse.io/ai-agents/tools/dataforseo.md) + [GitHub](https://agentuse.io/ai-agents/tools/github.md) + [Slack](https://agentuse.io/ai-agents/tools/slack.md)
- Expected output: Migration evidence artifact

## Implementation guidance for coding agents

1. Inspect the project for its installed or pinned AgentUse version and existing `.agentuse` files.
2. Load the version-matched AgentUse core, creator, and tester guidance before authoring an agent.
3. Verify which MCP servers, CLIs, APIs, SDKs, credentials, and resource IDs actually exist. Do not infer capabilities from this page.
4. Choose the smallest connection and permission set that can complete the job. Prefer read-only access for research and reporting.
5. Mechanically approval-gate consequential writes such as posting, sending, deleting, merging, or changing external records.
6. Test manually before adding a schedule or verified webhook trigger.

## Operating boundary

Safe automatic work: Validate the approved request scope, paginate DataForSEO responses to the declared cap, record expected and collected counts, final cursor or offset, errors, task status, and reported cost, then normalize complete sources and save a cited review artifact.

Never without the configured approval path: Expose API credentials, exceed the approved request or spend ceiling, silently advance a partial result as the baseline, modify a website, or claim causation from before-and-after movement alone.