acculynx-automation

Overview of acculynx-automation skill

What acculynx-automation is for

acculynx-automation is a Claude skill for automating Acculynx work through Composio’s Rube MCP server. It is not a standalone Acculynx API wrapper; it guides an agent to discover the current Acculynx tool schemas with RUBE_SEARCH_TOOLS, verify an active Acculynx connection with RUBE_MANAGE_CONNECTIONS, and then execute the appropriate Rube MCP tools for the requested workflow.

Use this skill when you want an AI agent to handle Acculynx operations through the Composio toolkit while reducing schema guesswork and stale-tool errors.

Best-fit users and workflows

The acculynx-automation skill is best for teams already using Claude or another MCP-capable client and needing repeatable Acculynx task execution. Good fits include roofing, construction, CRM, or operations teams that want to automate structured Acculynx actions such as looking up records, preparing workflow steps, or coordinating tool calls through Rube.

It is especially useful when the exact available Acculynx actions may change, because the skill’s core behavior is to search tools first instead of assuming fixed parameters.

Key differentiator: schema discovery first

The main practical value is the required discovery pattern. Before acting, the agent should call RUBE_SEARCH_TOOLS for the user’s specific Acculynx use case. That returns available tool slugs, input schemas, suggested execution plans, and pitfalls. This makes acculynx-automation more reliable than a plain prompt that asks the model to “use Acculynx,” because it anchors execution to the live Rube MCP tool catalog.

Adoption requirements to check first

Before installing or relying on acculynx-automation for Workflow Automation, confirm three things: your client supports MCP, Rube MCP is configured at https://rube.app/mcp, and your Acculynx connection can be made ACTIVE through RUBE_MANAGE_CONNECTIONS with toolkit acculynx. If any of those are missing, the skill can still describe a workflow, but it cannot execute Acculynx operations.

How to Use acculynx-automation skill

acculynx-automation install context

Install the skill from the Composio skill collection with your skill manager, for example:

npx skills add ComposioHQ/awesome-claude-skills --skill acculynx-automation

Then configure Rube MCP in your MCP-capable client by adding the server endpoint:

https://rube.app/mcp

The upstream skill does not include helper scripts or extra reference files, so the important source to inspect is composio-skills/acculynx-automation/SKILL.md. That file defines the required MCP dependency, the connection workflow, and the “search tools first” rule.

Inputs the skill needs from you

For high-quality acculynx-automation usage, give the agent more than a broad command. Include the business task, target Acculynx object or workflow area, known identifiers, constraints, and whether the agent should only plan or actually execute.

Weak prompt: “Update Acculynx.”

Stronger prompt: “Use acculynx-automation to find the current Acculynx tools for updating a job record. First confirm the acculynx connection is ACTIVE. Then identify the required schema for updating job status. Do not execute the update until you show me the required fields and ask for confirmation.”

This works better because it tells the skill what tool discovery should search for, how cautious execution should be, and what confirmation boundary to respect.

Recommended workflow for first run

Start with a connection check, not the business action. Ask the agent to verify that RUBE_SEARCH_TOOLS is available, then use RUBE_MANAGE_CONNECTIONS with toolkit acculynx. If the connection is not active, follow the returned authorization link and rerun the check.

After that, run discovery for the exact job-to-be-done, such as “search Acculynx customers,” “retrieve project details,” or “prepare an Acculynx update workflow.” The agent should reuse the returned session ID where appropriate, inspect tool schemas, and only then call execution tools.

Practical prompt pattern

A reliable acculynx-automation guide prompt looks like this:

• “Search Rube tools for this exact Acculynx task: [task].”
• “Check whether the acculynx toolkit connection is ACTIVE.”
• “Show the tool slug, required inputs, optional inputs, and any pitfalls.”
• “If execution is safe, perform [action]; otherwise ask me for missing fields.”
• “Do not invent field names; use the schema returned by RUBE_SEARCH_TOOLS.”

This pattern reduces the two biggest failure modes: acting before authentication is ready and hallucinating Acculynx tool parameters.

acculynx-automation skill FAQ

Is acculynx-automation enough without Rube MCP?

No. The skill depends on Rube MCP and the Composio Acculynx toolkit. Without RUBE_SEARCH_TOOLS and RUBE_MANAGE_CONNECTIONS, it becomes only an instruction document. Install it only if your environment can expose Rube MCP tools to the agent.

How is this better than an ordinary Acculynx prompt?

A normal prompt may rely on the model’s memory of APIs or integrations. The acculynx-automation skill tells the agent to discover live tool schemas first, check the active connection, and execute through Rube MCP. That matters for Acculynx automation because tool names, required fields, and auth state are operational details that generic prompts often miss.

Is this suitable for beginners?

Yes, if the beginner is comfortable setting up an MCP server and following an authentication link. The skill’s workflow is simple, but it assumes the user understands that tool calls can affect real Acculynx data. New users should start in a read-only or confirmation-first mode: discover tools, inspect schemas, and ask before writing changes.

When should I not use this skill?

Do not use acculynx-automation if you need offline documentation only, direct custom API development, unsupported Acculynx actions outside the Composio toolkit, or fully autonomous writes without review. It is also a poor fit if your organization cannot authorize Acculynx access through Rube MCP.

How to Improve acculynx-automation skill

Improve acculynx-automation results with clearer task scope

The skill performs best when the requested use case is specific. Replace “manage a claim” with “discover tools for retrieving an Acculynx job by ID and checking its current status.” Replace “sync customers” with “find whether Acculynx tools can search customers by email and return customer IDs.” Specific use cases produce better RUBE_SEARCH_TOOLS results and cleaner execution plans.

Add confirmation rules for write operations

For safer workflow automation, explicitly separate discovery, planning, and execution. Tell the agent which actions require approval, such as creating records, updating job status, modifying contacts, or sending data to another system. A useful rule is: “Read operations may run after schema discovery; write operations require a summary of intended changes and my approval.”

Watch for common failure modes

The most common issues are inactive Acculynx connection, skipped tool discovery, invented schema fields, and overbroad prompts. If output looks uncertain, ask the agent to rerun RUBE_SEARCH_TOOLS with a narrower use case and quote the required fields from the returned schema before continuing.

Iterate after the first output

After the first acculynx-automation run, improve the prompt with what the agent learned: the tool slug, required identifiers, missing fields, and any returned pitfalls. For repeated workflows, save a reusable prompt template that includes connection check, schema discovery, confirmation boundary, and expected output format. This turns the skill from a one-off automation helper into a more dependable Acculynx operating procedure.