breeze-automation
by ComposioHQbreeze-automation helps agents automate Breeze tasks through Composio Rube MCP by discovering current tool schemas, checking the Breeze connection, executing safely, and verifying results.
This skill scores 66/100, which means it is acceptable for directory listing but should be presented as a lightweight connector-oriented skill rather than a fully worked Breeze automation playbook. Directory users can understand when to use it and how to start, but should expect to rely on Rube MCP tool discovery for the actual Breeze schemas and task-specific execution details.
- Clear prerequisites identify the required Rube MCP server, RUBE_SEARCH_TOOLS availability, and an active Breeze connection via RUBE_MANAGE_CONNECTIONS.
- The skill gives an explicit operational pattern: discover tools first, check the Breeze connection, then execute using current schemas.
- The frontmatter and description are valid and triggerable for agents asked to automate Breeze tasks through Composio/Rube MCP.
- No support files, scripts, examples, or reference materials are included beyond SKILL.md, so users get limited evidence of tested Breeze-specific workflows.
- The workflow is mostly generic Rube MCP discovery guidance and depends on live tool schemas from RUBE_SEARCH_TOOLS, which means agents still need to infer the exact Breeze operation at runtime.
Overview of breeze-automation skill
What breeze-automation does
breeze-automation is a Claude skill for automating Breeze project-management tasks through Composio’s Rube MCP server. It is designed for agents that can call MCP tools, discover available Breeze actions, check authentication status, and then execute Breeze operations using the current tool schema rather than relying on hard-coded examples.
Best fit for Workflow Automation users
The breeze-automation skill is best for teams that already use Breeze and want AI-assisted Workflow Automation for repetitive project operations: finding tasks, updating records, coordinating project data, or building multi-step task-management flows. It is most useful when your agent has access to Rube MCP and you want a safer pattern for tool use: discover tools first, confirm the Breeze connection, execute, then verify the result.
Key differentiator: schema-first execution
The important design choice is that breeze-automation tells the agent to call RUBE_SEARCH_TOOLS before running any Breeze action. That matters because Composio tool schemas can change. Instead of guessing parameter names from memory, the agent should retrieve the current Breeze tool slugs, input fields, execution guidance, and known pitfalls at runtime.
Adoption requirements and limits
This is not a standalone Breeze API client. You need an MCP-capable client, Rube MCP configured at https://rube.app/mcp, and an active Breeze connection managed through RUBE_MANAGE_CONNECTIONS. The repository path contains only SKILL.md, so expect a compact operational instruction set rather than scripts, examples, or a full reference library.
How to Use breeze-automation skill
breeze-automation install and setup context
Install the skill in a compatible skills environment, for example:
npx skills add ComposioHQ/awesome-claude-skills --skill breeze-automation
Then add Rube MCP as a server in your client configuration:
https://rube.app/mcp
Before asking the agent to change Breeze data, verify that RUBE_SEARCH_TOOLS is available. Next, use RUBE_MANAGE_CONNECTIONS with toolkit breeze. If the connection is not ACTIVE, complete the returned authorization flow and re-check status before continuing.
Inputs the skill needs from you
A weak prompt is “update Breeze tasks.” A useful breeze-automation usage prompt gives the agent enough context to discover the right tools and avoid ambiguous edits:
- The Breeze object type: project, task, comment, person, status, or milestone if known
- The target records or filters: project name, task title, assignee, due date, status, tag
- The intended action: search, create, update, assign, comment, summarize, or verify
- Safety rules: ask before bulk edits, preview matches, do not close tasks without confirmation
- Output format: concise summary, table of changed records, or follow-up questions
Example prompt:
“Use breeze-automation to find all open Breeze tasks in the Website Redesign project assigned to Maya and due before Friday. First discover the current Breeze tools with RUBE_SEARCH_TOOLS, confirm the Breeze connection is active, show me the matching tasks, and ask for approval before changing dates or statuses.”
Practical workflow for reliable results
A dependable breeze-automation guide follows this sequence:
- Search tools with a specific use case, such as “find and update Breeze tasks by project and due date.”
- Reuse the returned session ID when checking connection and executing related actions.
- Confirm the Breeze toolkit connection is
ACTIVE. - Use the discovered tool schema exactly, including required fields and expected IDs.
- For destructive or broad changes, preview records first and request confirmation.
- After execution, verify the result with a follow-up read/search call.
This pattern is slower than a single direct command, but it reduces failed calls and accidental edits.
Repository files to read first
Start with composio-skills/breeze-automation/SKILL.md. There are no visible support folders such as rules/, resources/, references/, or scripts/ in the provided tree, so the skill’s behavior is concentrated in that one file. Pay special attention to the prerequisites, setup, tool discovery, and core workflow sections.
breeze-automation skill FAQ
Is breeze-automation useful without Rube MCP?
No. The breeze-automation skill depends on Rube MCP tools, especially RUBE_SEARCH_TOOLS and RUBE_MANAGE_CONNECTIONS. If your client cannot connect to MCP servers or call Rube tools, this skill will not provide practical automation.
How is this better than an ordinary Breeze prompt?
An ordinary prompt may cause the model to invent API fields or assume outdated tool names. breeze-automation gives the agent a required discovery step, so it should retrieve current Composio Breeze tool schemas before acting. That makes it better for live Workflow Automation where correctness depends on current tool definitions.
Is this suitable for beginners?
Yes, if the beginner already has an MCP-capable client and can complete the Breeze connection flow. The skill itself is short and operational. The main learning curve is not Breeze; it is understanding that the agent must discover tools, check authentication, and verify results instead of jumping straight to execution.
When should I not use this skill?
Do not use breeze-automation for offline planning, generic project-management advice, or Breeze workflows where you cannot authorize a live connection. Also avoid using it for large bulk updates unless your prompt requires previews, confirmation, and post-change verification.
How to Improve breeze-automation skill
Improve prompts with Breeze-specific constraints
The fastest way to improve breeze-automation results is to replace vague goals with operational constraints. Include the project name, task filters, fields to change, and what must not be touched. For example, “move overdue tasks” is risky; “list overdue open tasks in Project X, grouped by assignee, and ask before changing due dates” is much safer.
Prevent common failure modes
Common problems include missing Breeze authentication, skipped tool discovery, ambiguous record matches, and broad edits without confirmation. Counter these directly in the prompt: “Call RUBE_SEARCH_TOOLS first,” “confirm the Breeze connection is active,” “show matching records before updates,” and “verify the final state after execution.”
Iterate after the first output
Treat the first run as discovery when the task is complex. Ask the agent to return the available Breeze tools, required fields, and proposed execution plan before making changes. After the first result, refine filters such as assignee, status, date range, or project ID. This reduces accidental operations and helps the agent use the exact schema returned by Rube.
Improve the skill file for team use
If you maintain a local copy of breeze-automation, add team-specific examples to SKILL.md: approved project naming conventions, safe bulk-edit rules, required confirmation steps, and standard output formats. Keep the core rule intact: search tools first, use the current schema, check the Breeze connection, execute, and verify.