affinity-automation
by ComposioHQaffinity-automation helps agents run Affinity CRM operations through Composio Rube MCP. Use it to discover current tool schemas with RUBE_SEARCH_TOOLS, verify the Affinity connection, and follow safer CRM workflow steps before execution.
This skill scores 66/100, which makes it acceptable but limited for directory listing. Directory users get a usable MCP-triggering pattern for Affinity automation, especially around Rube discovery and connection setup, but should expect a relatively lightweight skill with little packaged reference material or concrete Affinity workflow depth.
- Valid skill frontmatter clearly declares the `rube` MCP requirement and describes the Affinity automation purpose.
- Prerequisites and setup steps explain that Rube MCP must be connected, Affinity must be ACTIVE via `RUBE_MANAGE_CONNECTIONS`, and `RUBE_SEARCH_TOOLS` should be called first.
- The core workflow pattern gives agents an actionable sequence for discovering tools, checking connection status, and executing Affinity operations through Composio/Rube.
- No support files, scripts, references, README, or install command are present, so installation/adoption guidance is limited to the SKILL.md text.
- The workflow relies on live Rube tool discovery for schemas and appears light on concrete Affinity-specific task examples, so agents may still need runtime interpretation.
Overview of affinity-automation skill
What affinity-automation is for
affinity-automation is a Claude skill for running Affinity CRM operations through Composio’s Rube MCP server. It is designed for tasks where an agent needs to discover the current Affinity tool schema, verify the user’s Affinity connection, and then execute CRM actions with less guesswork than a free-form prompt.
Best-fit users and jobs
The affinity-automation skill is a strong fit for CRM Operations, sales operations, investor relations, recruiting operations, and data teams that already use Affinity and want an AI agent to help with repeatable operational work. Typical jobs include finding the right Affinity tools, preparing list or organization updates, checking connection status, and guiding the agent through a safe “discover before execute” workflow.
What makes this skill different
The key differentiator is not a long library of hard-coded Affinity commands. The skill tells the agent to use RUBE_SEARCH_TOOLS first so it can retrieve current tool slugs, schemas, execution plans, and pitfalls from Rube MCP. That matters because MCP tool schemas can change; a prompt that assumes old parameter names may fail or write bad CRM data.
Important adoption requirements
Before installing or relying on affinity-automation, confirm that your AI client supports MCP and that Rube MCP is configured. The source skill requires mcp: [rube], an active Affinity connection through RUBE_MANAGE_CONNECTIONS, and access to the affinity toolkit. If you cannot connect Rube MCP or authorize Affinity, the skill will not perform useful work.
How to Use affinity-automation skill
affinity-automation install context
Install the skill from the Composio skill collection, then configure Rube MCP separately in your client. A typical install command is:
npx skills add ComposioHQ/awesome-claude-skills --skill affinity-automation
Then add https://rube.app/mcp as an MCP server in your client configuration. The upstream skill states that no API keys are needed for the endpoint itself, but you still need an active Affinity connection. First verify that RUBE_SEARCH_TOOLS is available, then use RUBE_MANAGE_CONNECTIONS with toolkit affinity and complete the returned authorization flow if the connection is not ACTIVE.
Inputs the skill needs from you
For useful affinity-automation usage, give the agent the business objective, the Affinity object type, the selection criteria, the intended action, and any safety rules. Weak prompt: “Update Affinity.” Stronger prompt: “Use affinity-automation to find Affinity tools for updating organizations. Check the Affinity connection first. I need to add companies from list X to list Y only if they are not already present. Do not overwrite existing owner, status, or notes fields. Show the discovered tool schema before executing.”
This improves output because the skill can map your operational goal to the current Rube tools instead of guessing whether the task involves people, organizations, lists, opportunities, or notes.
Recommended workflow
Start every run with tool discovery: ask the agent to call RUBE_SEARCH_TOOLS for the specific Affinity task, not just “Affinity operations.” Reuse the returned session where possible, inspect the discovered schema, then check the Affinity connection with RUBE_MANAGE_CONNECTIONS. Only after that should the agent call execution tools. For write operations, ask for a brief preflight summary: target records, fields to change, records to skip, and rollback or review limitations.
Repository files to read first
This skill is compact: the main file to inspect is composio-skills/affinity-automation/SKILL.md. There are no bundled scripts, references, rules, or README files in the current repository preview, so most implementation detail lives in the skill text and in the live Rube MCP tool discovery response. For install decisions, read the prerequisites, setup, tool discovery, and core workflow sections first.
affinity-automation skill FAQ
Is affinity-automation for CRM Operations?
Yes. affinity-automation for CRM Operations is most useful when the team needs a consistent agent workflow for Affinity tasks: discover tools, confirm connection, inspect schema, then act. It is especially relevant when data quality matters and the agent should avoid inventing fields or using stale API assumptions.
How is this better than an ordinary prompt?
An ordinary prompt may say “use Affinity,” but it may not force the agent to discover current Rube tool schemas. The affinity-automation skill encodes that operational discipline. Its value is highest when schemas, tool names, or required fields are uncertain, because it pushes the agent toward live tool discovery before execution.
Can beginners use this skill?
Beginners can use the affinity-automation skill if they understand their Affinity data model and have help configuring MCP. The hard part is usually not the skill text; it is knowing which records should be touched and what should never be changed. New users should start with read-only discovery and small test updates before allowing broad writes.
When should I not use it?
Do not use this skill if you do not have Rube MCP access, cannot authorize Affinity, or need a standalone script that runs without an AI client. It is also a poor fit for vague cleanup requests such as “fix our CRM” unless you can define target objects, matching rules, field rules, and review steps.
How to Improve affinity-automation skill
Improve affinity-automation prompts with operational rules
The fastest way to improve affinity-automation results is to write prompts like an operations runbook. Include the Affinity object, record source, match key, allowed fields, forbidden fields, duplicate handling, and whether the agent may execute or must only plan. Example: “Find tools for Affinity list membership. Use company domain as the match key. Add missing organizations to the named list. If multiple matches exist, skip and report. Do not create new organizations.”
Common failure modes to prevent
The main risks are skipped tool discovery, inactive connections, stale schema assumptions, ambiguous record matching, and overly broad write operations. Prevent these by requiring RUBE_SEARCH_TOOLS first, asking the agent to show the discovered schema, confirming ACTIVE connection status, and limiting the first execution to a small batch or dry-run style plan when available.
Iterate after the first output
After the first agent response, do not just approve execution. Check whether the selected Rube tools match the task, whether required fields are populated, and whether the agent identified pitfalls from the discovery response. If the plan is too broad, narrow the segment: “only records updated in the last 30 days,” “only organizations with a domain,” or “only list members from this source.”
Extend the skill for your team
Teams that use affinity-automation repeatedly can improve it by adding local guidance outside the upstream file: approved Affinity field names, naming conventions, list ownership rules, duplicate-resolution policy, and examples of safe prompts. Keep those additions separate from secrets, and update them when your Affinity workspace or Composio tool behavior changes.