dadata-ru-automation
by ComposioHQdadata-ru-automation helps agents run Dadata Ru workflows via Composio Rube MCP, requiring tool discovery, active dadata_ru connection checks, and schema-based execution for address, company, bank, and data-quality automation.
This skill scores 66/100, which means it is acceptable for directory listing but should be presented as a lightweight MCP/toolkit wrapper rather than a complete task playbook. Directory users get enough information to understand when to install it and how an agent should begin using Dadata Ru through Rube MCP, but the lack of concrete task examples and supporting files limits confidence and reuse.
- Defines a clear activation context: automate Dadata Ru operations through Composio's Dadata Ru toolkit via Rube MCP.
- Includes operational prerequisites and setup steps, including connecting Rube MCP, using `RUBE_MANAGE_CONNECTIONS`, and confirming an ACTIVE `dadata_ru` connection.
- Strongly instructs agents to call `RUBE_SEARCH_TOOLS` first, which reduces schema guesswork and improves safe tool execution.
- No support files, scripts, references, or examples beyond the SKILL.md, so adoption depends heavily on Rube MCP returning useful current schemas.
- Workflow guidance appears generic for the Dadata Ru toolkit rather than documenting concrete Dadata tasks, inputs, outputs, or edge cases.
Overview of dadata-ru-automation skill
What dadata-ru-automation does
dadata-ru-automation is a Claude skill for running Dadata.ru workflows through Composio’s Rube MCP toolkit. It is designed for agents that need to discover current Dadata Ru tool schemas, confirm an active connection, and then execute structured operations such as Russian address, company, bank, contact, or data-quality tasks without guessing API parameters.
Best fit for Workflow Automation teams
This skill is most useful when Dadata.ru is part of a repeatable business process: CRM enrichment, form normalization, lead cleanup, logistics address validation, counterparty lookup, or internal data operations. The main value of the dadata-ru-automation skill is not a static prompt; it enforces the correct MCP sequence: search tools first, verify connection state, then run the appropriate Dadata Ru action.
Key differentiator: schema discovery first
Dadata integrations can break when tool names, input fields, or execution plans change. This skill explicitly requires RUBE_SEARCH_TOOLS before execution, so the agent works from current Composio schemas instead of stale assumptions. That makes dadata-ru-automation for Workflow Automation a better fit than a generic “use Dadata” prompt when reliability matters.
What to check before installing
The repository path contains a single SKILL.md and no helper scripts, references, or bundled examples. That keeps the skill lightweight, but it also means your team should bring its own business rules: which entities to process, acceptable match confidence, locale expectations, error handling, and whether results should update a database, spreadsheet, CRM, or ticket.
How to Use dadata-ru-automation skill
dadata-ru-automation install and MCP setup
Install the skill from the Composio skill collection:
npx skills add ComposioHQ/awesome-claude-skills --skill dadata-ru-automation
Then configure Rube MCP in your client by adding:
https://rube.app/mcp
Before using the skill, confirm that RUBE_SEARCH_TOOLS is available. The Dadata Ru connection must also be active through RUBE_MANAGE_CONNECTIONS with toolkit dadata_ru. If the connection is not active, follow the returned authorization link and re-check status before asking the agent to process data.
Inputs the skill needs from you
For strong dadata-ru-automation usage, give the agent more than “clean this data.” Include:
- The Dadata Ru task type, such as address suggestion, party lookup, bank lookup, phone/email/name cleanup, or enrichment.
- Sample input rows or field names.
- Desired output fields and format.
- Whether the workflow is one-off, batch, or part of a larger automation.
- Rules for ambiguous matches, missing values, and low-confidence results.
- Destination system, if results need to be written back somewhere.
A weak prompt is: “Use Dadata for these addresses.”
A stronger prompt is: “Use dadata-ru-automation to normalize these Russian delivery addresses. First discover current Dadata Ru tools with Rube, then return standardized address, postal code, region, city, geo coordinates if available, confidence/quality fields, and a list of rows that need manual review. Do not overwrite source values.”
Practical workflow for first run
Start by reading composio-skills/dadata-ru-automation/SKILL.md. It contains the required operating pattern:
- Call
RUBE_SEARCH_TOOLSwith a specific use case, not a broad one. - Use the returned tool slugs, schemas, and pitfalls.
- Check the Dadata Ru connection state through Rube.
- Execute the chosen tool with schema-compliant inputs.
- Inspect outputs before applying updates to downstream systems.
For example, use “Dadata Ru company lookup by INN and return official name, status, address, and management fields” instead of “Dadata Ru operations.” Specific discovery queries produce more relevant tool recommendations and reduce failed calls.
Tips for better prompts and safer automation
Ask the agent to show the selected tool schema before execution when you are testing. For production-like runs, require a dry-run table with input, normalized output, confidence indicators, and proposed action. If the workflow writes to another system, separate lookup from write-back: first collect Dadata results, then ask for an update plan, then approve execution.
dadata-ru-automation skill FAQ
Is dadata-ru-automation only for Russian data?
Yes, the skill targets the Dadata Ru toolkit, which is most relevant for Russian addresses, organizations, banks, and personal/contact data formats. It is not a general international data-cleaning skill. If your dataset is mostly outside Dadata.ru’s coverage, use a broader enrichment or validation workflow instead.
How is this better than an ordinary prompt?
A normal prompt may invent endpoint names or assume old request fields. dadata-ru-automation tells the agent to use Rube MCP discovery first, then follow the live schema returned by Composio. That matters when automating real workflows because the agent can adapt to available tools instead of relying on memory.
Do beginners need to know the Dadata API?
Not deeply, but beginners should understand the business goal and the shape of their data. The skill can discover tools and schemas, but it cannot decide your quality thresholds, what to do with duplicates, or whether a returned organization/status is acceptable for your process. Treat it as an automation guide, not a replacement for data-governance rules.
When should I not use this skill?
Do not use dadata-ru-automation if you cannot connect Rube MCP, cannot activate the Dadata Ru toolkit, or only need offline text cleanup with no Dadata lookup. It is also a poor fit for high-volume unattended updates until you have tested rate limits, error handling, and review rules on a small batch.
How to Improve dadata-ru-automation skill
Make dadata-ru-automation inputs more explicit
The easiest way to improve results is to provide a precise task contract. Include column names, examples, required fields, acceptable null behavior, and the final output destination. For instance, “normalize raw_address into postal_code, region, city, street, house, geo_lat, geo_lon, and qc” gives the agent a concrete schema to map against discovered Dadata tools.
Add review rules for ambiguous matches
Common failure modes include multiple possible organizations, partial addresses, outdated company records, transliteration issues, and missing identifiers. Tell the agent how to handle them: “If confidence is low, mark needs_review=true,” “Do not choose between multiple parties without INN/KPP,” or “Preserve the original user-entered value in source_value.”
Iterate after the first tool discovery
After the first RUBE_SEARCH_TOOLS call, ask the agent to summarize available tool slugs, required inputs, optional fields, and known pitfalls before execution. This turns the dadata-ru-automation guide into an auditable plan and helps you catch mismatches early, such as trying to enrich companies when the data only contains names without INN or address context.
Extend the skill for your own workflow
Because the upstream skill is intentionally minimal, teams can improve it by adding local examples, batch templates, validation checklists, or post-processing rules. Useful additions include sample prompts for address normalization, party lookup by INN, bank lookup by BIK, CRM enrichment, and CSV review workflows. Keep those extensions separate from credentials and avoid hard-coding schemas that should be discovered through Rube.