daffy-automation

Overview of daffy-automation skill

What daffy-automation is for

daffy-automation is a Claude skill for automating Daffy-related actions through Composio’s Rube MCP server. It is designed for users who want an agent to discover available Daffy tools, check authentication, and execute Daffy workflows using current tool schemas instead of relying on hard-coded assumptions.

The practical job-to-be-done is not “write a generic Daffy prompt.” It is to make Claude operate through Rube MCP safely: search tools first, confirm the Daffy connection is active, inspect the returned schema, and then call the right tool with valid inputs.

Best-fit users and workflows

This skill is a good fit if you already use Claude with MCP and want workflow automation around Daffy via Composio. It is most useful for repeatable operational tasks where the exact Daffy tool name, required fields, or execution plan may change over time.

Typical users include automation builders, operations teams, and developers wiring charitable giving workflows into a broader agent process. The key value is schema-aware execution: the skill tells the agent to ask Rube what tools are currently available before attempting an action.

Main differentiator: tool discovery first

The strongest reason to install the daffy-automation skill is its disciplined workflow pattern. It explicitly requires RUBE_SEARCH_TOOLS before execution, then RUBE_MANAGE_CONNECTIONS to verify the daffy toolkit connection. That matters because MCP tool schemas can change, and a stale prompt can fail silently or submit incomplete data.

This skill is intentionally lightweight. It does not include helper scripts, reference files, or custom business rules. Its usefulness depends on your Rube MCP setup and the quality of the Daffy task description you provide.

How to Use daffy-automation skill

daffy-automation install and setup context

Install the skill from the Composio skill collection:

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

Then configure Rube MCP in your client by adding:

https://rube.app/mcp

Before expecting the skill to work, confirm that the MCP server exposes RUBE_SEARCH_TOOLS. Next, use RUBE_MANAGE_CONNECTIONS with toolkit daffy. If the connection is not ACTIVE, follow the returned authentication link and complete the Daffy connection flow. Do not run Daffy operations until the connection status is active.

Inputs the skill needs from you

A weak request is: “Use Daffy.” A better request gives the agent a goal, constraints, and the information it should discover before acting.

Example prompt:

Use the daffy-automation skill. First call RUBE_SEARCH_TOOLS for the specific Daffy task I describe, then check the daffy connection with RUBE_MANAGE_CONNECTIONS. I want to [describe the exact Daffy operation]. Do not execute until you show me the selected tool, required fields, and any missing inputs.

Useful details to provide include the intended Daffy action, relevant account or organization context, whether the agent should only draft or actually execute, approval requirements, and any fields you already know. This helps the agent turn a broad workflow automation goal into a valid MCP tool call.

Recommended daffy-automation usage workflow

Use this order for reliable daffy-automation usage:

1. Ask Claude to invoke the skill for a specific Daffy outcome.
2. Require RUBE_SEARCH_TOOLS with a use case matching your task, not a vague “Daffy operations” query.
3. Review the returned tool slugs, schemas, execution plan, and pitfalls.
4. Confirm the Daffy connection through RUBE_MANAGE_CONNECTIONS.
5. Ask Claude to map your available data to the required schema.
6. Approve execution only after the agent identifies missing or risky fields.

This workflow is slower than a one-line prompt, but it reduces failed calls and prevents the agent from guessing parameters.

Repository files to read first

The skill currently consists of SKILL.md only. Read it before installation if you need to verify prerequisites or understand the intended execution order. The most important sections are Prerequisites, Setup, Tool Discovery, and Core Workflow Pattern.

Because there are no bundled scripts, rules, resources, or metadata files, your own prompt and MCP environment carry most of the operational detail.

daffy-automation skill FAQ

Is daffy-automation beginner-friendly?

It is beginner-friendly only if your Claude client already supports MCP and you are comfortable authorizing a Daffy connection through Rube. The skill gives a clear sequence, but it does not hide the need to inspect tool schemas or approve live operations.

If you are new to MCP, start by confirming that RUBE_SEARCH_TOOLS works before trying any Daffy task.

How is this better than an ordinary prompt?

An ordinary prompt may ask Claude to “do something in Daffy,” but it may invent tool names or assume outdated fields. The daffy-automation skill centers the workflow on live tool discovery through Rube MCP. That makes it more suitable for Workflow Automation where reliability depends on current schemas and authenticated tool access.

When should I not use this skill?

Do not use it if you do not have Rube MCP available, cannot connect the Daffy toolkit, or only need general advice about Daffy rather than tool execution. It is also a poor fit for unsupervised high-risk actions unless you add your own approval gates, logging, and validation steps.

Does the skill include ready-made Daffy workflows?

No. It provides a reliable pattern for discovering and executing Daffy tools, not a library of prebuilt workflows. That is a tradeoff: less out-of-the-box automation, but better adaptability to the current Composio Daffy toolkit schema.

How to Improve daffy-automation skill

Improve prompts before running daffy-automation

The biggest quality gain comes from writing a precise operating prompt. Include the task, desired outcome, execution permissions, known data, and review checkpoint.

Stronger prompt pattern:

Use daffy-automation for this task: [specific Daffy goal]. Search tools first with that exact use case. Check the daffy connection. Then summarize the matching tool, required schema, missing fields, and risks. Wait for my confirmation before calling the execution tool.

This prevents premature execution and forces schema mapping before action.

Common failure modes to avoid

The most common failure is skipping RUBE_SEARCH_TOOLS and relying on assumed tool names. Another is treating an inactive Daffy connection as a prompt problem instead of an authentication problem. A third is giving the agent an outcome without enough field-level context, which leads to repeated clarification loops.

If a call fails, ask Claude to re-run tool discovery, compare the attempted payload against the returned schema, and identify which required field or permission blocked the workflow.

Add your own operating rules

For production use, extend the daffy-automation guide with local rules such as “never execute without human approval,” “log selected tool slug and payload before calling,” or “redact sensitive donor/account data in summaries.” These rules are not included upstream, but they materially improve safety and auditability.

You can also create reusable prompt snippets for common Daffy tasks so users provide consistent inputs each time.

Iterate after the first output

After the first agent response, do not only check whether it “looks right.” Verify that it searched tools, used the active daffy connection, selected a tool based on the returned schema, and asked for missing required fields. If any step is absent, ask it to restart from discovery.

That iteration discipline is what turns daffy-automation from a simple skill install into a dependable workflow automation pattern.