tomba-automation
by ComposioHQtomba-automation is a Claude skill for running Tomba lead research through Composio Rube MCP. It guides agents to install Rube MCP, confirm the Tomba connection, discover current tool schemas first, and execute email finding or verification workflows safely.
This skill scores 66/100, so it is acceptable to list but should be presented as a lightweight connector workflow rather than a rich Tomba playbook. Directory users can understand when to trigger it and how to start through Rube MCP, but they should expect the actual Tomba operation details to come from runtime tool discovery rather than from repository-provided examples or reference material.
- Valid frontmatter declares the required `rube` MCP and a concise trigger: automate Tomba tasks via Rube MCP.
- Prerequisites and setup are explicit: connect Rube MCP, use `RUBE_MANAGE_CONNECTIONS` for toolkit `tomba`, and confirm an ACTIVE connection.
- The skill repeatedly instructs agents to call `RUBE_SEARCH_TOOLS` first, which reduces schema guesswork and supports current tool execution.
- No support files, install command, or examples beyond the SKILL.md; adoption depends on knowing how to configure MCP clients and Rube.
- The workflow appears mostly generic to Rube/Tomba discovery and connection management, so users get limited task-specific Tomba guidance before runtime tool discovery.
Overview of tomba-automation skill
What tomba-automation does
tomba-automation is a Claude skill for running Tomba lead research and email-finding workflows through Composio’s Rube MCP. Instead of guessing tool names or hard-coding an old schema, the skill instructs the agent to discover the current Tomba tools first, confirm the Tomba connection, and then execute the right action with the latest available inputs.
Best fit for Lead Research workflows
The best use case is structured lead research: finding company contacts, validating email-related data, enriching prospect records, or automating repeatable Tomba tasks from a clear target list. It is useful when you want Claude to operate Tomba through MCP tools rather than only draft search queries, spreadsheet formulas, or manual research instructions.
Key differentiator: tool discovery first
The main value of the tomba-automation skill is its “search tools first” pattern. Composio tool schemas can change, so the skill pushes the agent to call RUBE_SEARCH_TOOLS before execution. That reduces failed calls, missing fields, and outdated assumptions about available Tomba actions.
What to check before installing
Install this skill only if your Claude environment can use MCP and you can add the Rube MCP server. You also need an active Tomba connection through RUBE_MANAGE_CONNECTIONS. If you only need a written lead research plan without live tool execution, a normal prompt may be enough.
How to Use tomba-automation skill
tomba-automation install and setup
Install the skill from the Composio skills repository:
npx skills add ComposioHQ/awesome-claude-skills --skill tomba-automation
Then configure Rube MCP in your client by adding:
https://rube.app/mcp
Before using the skill, verify that RUBE_SEARCH_TOOLS is available. Next, call RUBE_MANAGE_CONNECTIONS with toolkit tomba. If the connection is not ACTIVE, follow the returned authorization link and confirm the status before asking the agent to run any Tomba workflow.
Inputs the skill needs from you
A weak request like “find leads with Tomba” gives the agent too little to work with. Provide the target, desired output, limits, and what should happen if data is missing.
Stronger prompt:
Use tomba-automation for Lead Research. Find likely work emails for 25 marketing leaders at B2B SaaS companies in Germany. Use Tomba via Rube MCP, discover current tools first, confirm the Tomba connection, and return a table with company, person, role, domain, email result, confidence or verification status if available, and notes for records that need manual review.
This works better because it tells the skill which Tomba task to discover, how many records to process, what fields matter, and how to handle uncertainty.
Practical tomba-automation usage workflow
A reliable tomba-automation usage flow is:
- Ask Claude to use the
tomba-automationskill. - State the exact Tomba job: domain search, email finding, verification, enrichment, or another lead research task.
- Require
RUBE_SEARCH_TOOLSfirst with your specific use case. - Confirm the
tombaconnection throughRUBE_MANAGE_CONNECTIONS. - Have the agent choose the current tool slug and schema from discovery results.
- Run a small test batch before processing a large list.
- Review failures, missing fields, rate limits, or low-confidence records before continuing.
This staged approach prevents wasting credits or time on a large run with the wrong schema or incomplete input data.
Repository files to read first
The repository path is composio-skills/tomba-automation, and the important file is SKILL.md. Read it for the prerequisites, Rube MCP setup, tool discovery pattern, and core workflow. There are no extra scripts, rules, resources, or reference folders in the current skill package, so your operating quality depends heavily on your prompt and on live RUBE_SEARCH_TOOLS results.
tomba-automation skill FAQ
Is tomba-automation beginner friendly?
It is beginner friendly if you already use Claude with MCP tools, but it is not a no-setup lead database. The skill assumes Rube MCP is connected and that you can authorize the Tomba toolkit. Beginners should start with one domain or a small contact list, then inspect the tool discovery output before running bigger jobs.
How is this better than an ordinary prompt?
A normal prompt can describe how to research leads, but it cannot safely infer the current Composio Tomba tool schema. The tomba-automation skill gives Claude an execution pattern: discover tools, check connection, use current schemas, then run the workflow. That makes it better for real tool calls than a generic “use Tomba” instruction.
When should I not use this skill?
Do not use it when you lack permission to process the prospect data, cannot connect Rube MCP, do not have access to Tomba through Composio, or only need strategic advice. It is also a poor fit for broad web scraping jobs that are not actually Tomba operations.
Does it replace CRM or sales engagement tools?
No. tomba-automation is best treated as an execution layer for Tomba tasks inside an AI workflow. You may still need a CRM, spreadsheet, enrichment pipeline, or sales engagement platform to store, deduplicate, sequence, and govern the resulting lead records.
How to Improve tomba-automation skill
Improve tomba-automation results with better prompts
The most important improvement is giving the agent operational constraints, not just a goal. Include the target market, record count, accepted roles, required output columns, confidence threshold, deduplication rules, and whether to stop on authentication, quota, or schema errors.
Example refinement:
If Tomba returns multiple possible emails, keep the highest-confidence result, include alternates in a notes column, and flag anything below the confidence threshold for manual review instead of treating it as usable.
Common failure modes to prevent
The main failures are skipping RUBE_SEARCH_TOOLS, running before the tomba connection is active, assuming an old input schema, processing too many records before testing, and accepting incomplete lead data as final. Add explicit instructions to discover tools first, run a sample, and summarize any tool errors or missing fields.
Iterate after the first output
After the first run, do not just ask for “more.” Review which rows failed and why. Then ask Claude to adjust the next run: narrow the role list, provide missing domains, change the output format, or retry only records with recoverable errors. This makes the tomba-automation guide more efficient for real Lead Research because each batch improves the next.
Add local workflow notes if your team reuses it
If your team installs this skill for repeated lead research, document your preferred prompt template, approved output columns, compliance requirements, naming conventions, and handoff destination. The upstream skill is intentionally lightweight, so team-specific operating rules are the fastest way to make tomba-automation consistent across users.