conveyor-automation
by ComposioHQconveyor-automation is a Claude skill for using Composio Rube MCP to automate Conveyor tasks. It guides agents to verify the Conveyor connection, search live tool schemas with RUBE_SEARCH_TOOLS, and plan safe workflow actions before execution.
Score: 64/100. This is an acceptable but limited listing candidate: directory users can understand when to use it and how an agent should start safely via Rube MCP, but the repository evidence suggests it is mostly a discovery-and-connection pattern rather than a rich Conveyor automation playbook. It is useful for users already relying on Rube/Composio, with clear cautions about runtime dependency and limited concrete workflow detail.
- Valid skill metadata clearly identifies the trigger area: automating Conveyor operations through Composio's Conveyor toolkit via Rube MCP.
- Prerequisites and setup steps are explicit, including the need for Rube MCP, an active Conveyor connection, and use of RUBE_MANAGE_CONNECTIONS.
- The skill gives agents an important operational constraint: always call RUBE_SEARCH_TOOLS first to retrieve current tool schemas before executing workflows.
- No support files, scripts, examples, or repository-local references are included beyond SKILL.md, so adoption depends heavily on Rube MCP tool discovery at runtime.
- The available evidence shows a generic Conveyor/Rube workflow pattern rather than concrete end-to-end Conveyor task recipes, which may leave users guessing for specific operations.
Overview of conveyor-automation skill
What conveyor-automation does
conveyor-automation is a Claude skill for running Conveyor workflow tasks through Composio’s Rube MCP server. Its main value is not a fixed script; it gives the agent an operating pattern: connect to Rube MCP, confirm an active Conveyor toolkit connection, search the live tool catalog, then execute Conveyor actions using the current schemas returned by Rube.
This matters because Conveyor tool names and input fields can change. The skill explicitly tells the agent to call RUBE_SEARCH_TOOLS before acting, which reduces brittle prompts and outdated assumptions.
Best-fit users and use cases
The conveyor-automation skill is best for teams already using Conveyor and wanting Claude or another skill-compatible agent to operate Conveyor through Composio. It fits workflow automation tasks such as checking available Conveyor operations, preparing execution plans, running authorized actions, and troubleshooting connection status before a run.
It is most useful when you need an agent to interact with Conveyor repeatedly and safely, rather than asking a one-off generic prompt like “use Conveyor to do X.”
What differentiates this skill
The differentiator is its tool-discovery-first workflow. Instead of hardcoding Conveyor actions, the skill requires the agent to retrieve live tool slugs, schemas, recommended plans, and pitfalls from Rube MCP. That makes it better suited for environments where the Conveyor API surface may evolve or where tool availability depends on the authenticated connection.
The tradeoff is that this is a thin integration skill. It does not include extra scripts, examples, reference files, or domain-specific Conveyor playbooks beyond SKILL.md.
How to Use conveyor-automation skill
conveyor-automation install and setup context
Install the skill from the GitHub skill repository with your skill manager, for example:
npx skills add ComposioHQ/awesome-claude-skills --skill conveyor-automation
Then configure Rube MCP in your client by adding the MCP server endpoint:
https://rube.app/mcp
The skill requires Rube MCP and expects RUBE_SEARCH_TOOLS to be available. Before asking for any Conveyor action, use RUBE_MANAGE_CONNECTIONS with toolkit conveyor and confirm the connection is ACTIVE. If the connection is not active, follow the authentication link returned by Rube and retry the status check.
Inputs the skill needs for reliable usage
For strong conveyor-automation usage, give the agent a concrete Conveyor goal, the business constraint, and the expected output format. Avoid vague requests such as “automate Conveyor.” A better prompt is:
Use the conveyor-automation skill to find the current Rube MCP tools for Conveyor, verify my Conveyor connection, and create an execution plan for updating the target workflow. Do not run changes until you show the tool slug, required fields, risks, and confirmation question.
This improves output because the skill depends on live schema discovery. The agent needs to know the specific Conveyor task so RUBE_SEARCH_TOOLS can search for the right use case instead of returning a broad or irrelevant tool set.
Recommended workflow for first run
Start by reading composio-skills/conveyor-automation/SKILL.md; there are no supporting README.md, rules/, references/, or scripts in this skill directory, so the main file is the source of truth.
A safe first workflow is:
- Confirm
RUBE_SEARCH_TOOLSresponds. - Call
RUBE_MANAGE_CONNECTIONSfor toolkitconveyor. - If not active, complete the auth flow.
- Search tools with your exact Conveyor use case.
- Review returned schemas and execution plan.
- Ask the agent to perform a dry-run summary before invoking write actions.
- Execute only after confirming the target, fields, and expected side effects.
Prompt pattern that invokes the skill well
Use a prompt that separates discovery, planning, and execution:
Use conveyor-automation for Workflow Automation. First call
RUBE_SEARCH_TOOLSfor this exact task:[describe task]. Then check the Conveyor connection. Summarize the available tool slugs, required parameters, and possible pitfalls. If a write action is needed, stop and ask me for approval before execution.
This pattern keeps the agent aligned with the skill’s core instruction: search tools first, then act based on current schemas.
conveyor-automation skill FAQ
Is conveyor-automation only for Composio users?
Yes, practically. The skill is built around Composio’s Rube MCP and the Conveyor toolkit exposed through it. If your agent environment cannot connect to https://rube.app/mcp, cannot call RUBE_SEARCH_TOOLS, or cannot authenticate a Conveyor connection through RUBE_MANAGE_CONNECTIONS, this skill will not deliver its intended value.
How is it better than an ordinary Conveyor prompt?
An ordinary prompt may hallucinate tool names, assume stale parameters, or skip authentication checks. The conveyor-automation skill forces a safer sequence: discover current tools, verify connection status, inspect schema requirements, then run the appropriate operation. That is the main reason to install it instead of relying on natural-language instructions alone.
Is this conveyor-automation guide beginner-friendly?
It is beginner-friendly if you already understand MCP basics and have access to Conveyor. The skill’s own content is concise, so beginners should be ready to inspect the live output from RUBE_SEARCH_TOOLS. The agent will need to reason from returned schemas rather than follow a long built-in tutorial.
When should I not use this skill?
Do not use it for generic factory conveyor systems, physical robotics, PLC programming, or non-Composio automation. The name refers to Conveyor as a Composio toolkit target, not industrial conveyor-belt automation. Also avoid it when you need a fully packaged workflow with examples, tests, or helper scripts; this skill is a connection and execution pattern, not a complete application.
How to Improve conveyor-automation skill
Make conveyor-automation prompts more specific
The most common failure mode is asking for a broad action without enough context. Improve results by including the exact Conveyor object or workflow, whether the agent may make changes, what should be preserved, and what output you expect.
Weak:
Use Conveyor to fix my workflow.
Stronger:
Use conveyor-automation to inspect available Conveyor tools for modifying workflow routing. Verify the connection, list required fields, and propose the safest update plan. Do not execute write actions until I approve.
Use live schemas as the decision source
Because this skill depends on Rube MCP discovery, do not assume field names from memory or old documentation. After RUBE_SEARCH_TOOLS, have the agent restate the chosen tool slug, required inputs, optional inputs, and any known pitfalls returned by Rube. This makes the next action auditable and helps catch mismatches before execution.
Add guardrails before write operations
For production workflow automation, ask for a pre-execution checkpoint. The checkpoint should include target resource, intended mutation, rollback or recovery idea, missing inputs, and a yes/no confirmation request. This is especially important because the skill can bridge from conversation to real Conveyor actions once the connection is active.
Iterate after the first output
If the first plan is too broad, narrow the use case and rerun tool discovery with a better query. If the agent selects the wrong operation, paste the relevant schema result back into the conversation and ask it to compare alternatives. The conveyor-automation skill improves when you treat discovery output as working context, not as a hidden implementation detail.