googlesuper-automation
by ComposioHQgooglesuper-automation helps agents run Google Super operations via Rube MCP by searching live tool schemas first, checking the googlesuper connection, and executing with fewer guessed parameters.
This skill scores 64/100, which makes it acceptable but limited for directory listing. Directory users can understand that it is a Rube MCP-based wrapper for Google Super automation and get enough setup guidance to try it, but they should expect to rely on live tool discovery rather than repository-provided examples or detailed workflows.
- Valid skill frontmatter declares the Rube MCP requirement and names the googlesuper automation scope clearly.
- Provides actionable prerequisites and setup steps, including checking RUBE_SEARCH_TOOLS and activating the googlesuper connection through RUBE_MANAGE_CONNECTIONS.
- Emphasizes schema discovery before execution, which should reduce stale-tool assumptions for agents using Composio/Rube.
- No support files, scripts, references, README, or install command beyond the MCP endpoint instructions, so adoption relies entirely on the SKILL.md.
- The workflow is generic and depends on live RUBE_SEARCH_TOOLS discovery; it does not provide concrete Google Super task examples or known tool slugs in the repository evidence.
Overview of googlesuper-automation skill
What googlesuper-automation does
googlesuper-automation is a Claude skill for running Google Super operations through Composio’s Rube MCP server. Its main purpose is not to hard-code one Google workflow, but to make the agent discover the current Google Super tool schemas first, confirm connection status, and then execute the right Rube tool with less guessing.
Best fit for workflow automation users
The googlesuper-automation skill is best for users who already work with MCP-enabled assistants and want Google Super actions handled as part of broader Workflow Automation. It is most useful when the exact available tools or input fields may change, because the skill explicitly instructs the agent to call RUBE_SEARCH_TOOLS before execution instead of relying on stale assumptions.
What makes this skill different
The key differentiator is its tool-discovery-first pattern. Many automation prompts jump directly to an action; googlesuper-automation requires the agent to search available tools, inspect schemas, check the googlesuper connection, and only then run the task. That matters for reliability because Composio tool slugs, parameters, and recommended execution plans can vary over time.
Adoption considerations
This is a compact skill with a single SKILL.md and no bundled scripts, examples, or reference files. That keeps the install lightweight, but it also means users must be comfortable reading the Rube tool responses and giving the agent enough task context. It requires Rube MCP access and an active Google Super connection; without those, the skill cannot perform useful work.
How to Use googlesuper-automation skill
googlesuper-automation install context
Install the skill from the ComposioHQ skill repository in a compatible Claude skills environment:
npx skills add ComposioHQ/awesome-claude-skills --skill googlesuper-automation
Then add Rube MCP as a server in your client configuration using https://rube.app/mcp. The upstream skill notes that no API key is required for the endpoint itself, but you still need an active Google Super connection managed through Rube.
Required setup before running tasks
Before asking for an automation, confirm that RUBE_SEARCH_TOOLS is available. Then use RUBE_MANAGE_CONNECTIONS with toolkit googlesuper to verify the Google Super connection. If the returned status is not ACTIVE, follow the auth link provided by Rube and re-check the status before continuing.
The practical setup sequence is:
- Confirm Rube MCP responds.
- Search tools for the intended Google Super use case.
- Check or create the
googlesuperconnection. - Use the returned schema to plan the actual call.
- Execute only after required fields are known.
Turning a rough goal into a useful prompt
A weak googlesuper-automation usage prompt is: “Do my Google Super task.” It does not tell the agent what tools to search for, what data to use, or what success means.
A stronger prompt is:
“Use googlesuper-automation. First call RUBE_SEARCH_TOOLS for a Google Super workflow that [describe task]. Check that my googlesuper connection is active. Before executing, show me the selected tool slug, required fields, and any missing inputs. Use these details: [account/workspace/context], [records or target objects], [date range], [output format], [approval requirement].”
This works better because the skill depends on current schema discovery. The agent needs your concrete use case, target data, constraints, and whether it should pause for confirmation before making changes.
Files and repository path to inspect first
Start with composio-skills/googlesuper-automation/SKILL.md. There are no extra scripts/, references/, resources/, or README.md files in this skill, so the main value is in the required execution pattern. Pay special attention to the sections on prerequisites, setup, tool discovery, and the core workflow pattern.
googlesuper-automation skill FAQ
Is googlesuper-automation beginner-friendly?
It is beginner-friendly only if your Claude client already supports MCP and you are comfortable authorizing external tool connections. The skill gives a clear pattern, but it does not replace setup knowledge for Rube MCP or Google Super authentication.
How is it better than an ordinary prompt?
An ordinary prompt may ask the model to infer available actions from memory. The googlesuper-automation skill tells the agent to discover live tool schemas through RUBE_SEARCH_TOOLS, check connection state, and use the returned execution guidance. That reduces failures caused by outdated tool names or missing parameters.
When should I not use this skill?
Do not use it if you only need a written plan, a static explanation, or a one-off Google task outside Rube MCP. It is also a poor fit when your environment cannot connect to https://rube.app/mcp, when the googlesuper connection cannot be authorized, or when your organization does not allow agent-driven Google operations.
Does it include ready-made Google workflows?
No. The googlesuper-automation skill provides a reliable discovery and execution pattern for Google Super operations, not a library of prebuilt recipes. The exact action plan comes from the current Rube search results and the details you provide in your prompt.
How to Improve googlesuper-automation skill
Improve inputs for googlesuper-automation
The biggest improvement is to give the agent operational detail before tool discovery. Include the business goal, target Google Super object or area, exact records or filters, date ranges, required output, and whether destructive or external-facing actions need approval. This lets RUBE_SEARCH_TOOLS search for the right use case and helps the agent map your request to the returned schema.
Handle common failure modes
Common failures include inactive connections, skipped tool discovery, vague user goals, and attempts to call a tool with guessed fields. Prevent these by requiring the agent to summarize the chosen tool slug, required parameters, missing values, and expected side effects before execution. If a tool call fails, ask the agent to re-run discovery with the error context rather than retrying the same call blindly.
Iterate after the first output
After the first result, refine based on what Rube returned. If the output is too broad, add filters. If the wrong tool was selected, restate the use case and ask for a new RUBE_SEARCH_TOOLS query. If fields are ambiguous, provide examples of valid values from your workspace. For high-impact automations, run a read-only or preview-style step first when the discovered tools support it.
Extend the skill responsibly
Because the repository contains only SKILL.md, teams that use googlesuper-automation heavily may want to add local guidance outside the upstream skill: approved Google Super workflows, naming conventions, confirmation rules, rollback expectations, and examples of successful prompts. Keep those additions aligned with the core rule: always search tools first and treat the live schema as authoritative.