forcemanager-automation
by ComposioHQforcemanager-automation helps agents automate Forcemanager CRM tasks through Composio Rube MCP by verifying the connection, discovering current tool schemas, and using safer read-before-write workflows.
This skill scores 68/100, which means it is acceptable to list but best suited for users already using Rube MCP and Forcemanager. It gives agents a useful connection and tool-discovery pattern, but directory users should expect limited task-specific guidance and few repository artifacts beyond SKILL.md.
- Clear trigger and scope: use it for automating Forcemanager operations through Composio's Forcemanager toolkit via Rube MCP.
- Includes essential prerequisites and setup checks, including verifying RUBE_SEARCH_TOOLS and activating the Forcemanager connection with RUBE_MANAGE_CONNECTIONS.
- Emphasizes tool discovery before execution, which should help agents get current schemas instead of guessing tool inputs.
- Provides only a generic Rube MCP discovery/execution pattern; the excerpt shows no concrete Forcemanager task recipes or field-level examples.
- No support files, scripts, references, README, or install command beyond adding the Rube MCP endpoint, so adoption depends on already understanding MCP/Rube setup.
Overview of forcemanager-automation skill
What forcemanager-automation does
The forcemanager-automation skill helps an AI agent automate Forcemanager CRM tasks through Composio’s Rube MCP server. Its core value is not a fixed list of CRM actions; it teaches the agent to discover the current Forcemanager tool schemas first, verify the connection, then execute the right Rube tool with fewer schema guesses.
Best fit for CRM Operations teams
This is a strong fit for CRM Operations, RevOps, sales operations, or support teams that already use Forcemanager and want an agent to help with repeatable CRM work such as looking up records, updating entities, checking pipeline data, or coordinating account activity. It is most useful when your workflow depends on current Composio tool schemas rather than hardcoded API assumptions.
Key differentiator: schema discovery first
The most important behavior in this skill is mandatory tool discovery with RUBE_SEARCH_TOOLS before execution. That matters because MCP tool names, input fields, and execution plans can change. A generic prompt may invent parameters; the forcemanager-automation skill pushes the agent to retrieve current tool slugs, schemas, pitfalls, and recommended plans before acting.
Adoption requirements and limits
To use the forcemanager-automation skill, your client must support MCP and have Rube MCP connected at https://rube.app/mcp. You also need an active Forcemanager connection via RUBE_MANAGE_CONNECTIONS using the forcemanager toolkit. This skill does not include scripts, reference files, or offline examples, so the live Rube discovery step is essential.
How to Use forcemanager-automation skill
forcemanager-automation install context
Install the skill from the ComposioHQ skill collection if your environment supports Claude-style skills:
npx skills add ComposioHQ/awesome-claude-skills --skill forcemanager-automation
Then configure Rube MCP in your AI client with the server endpoint:
https://rube.app/mcp
Before asking the agent to run CRM work, confirm that RUBE_SEARCH_TOOLS is available. Next, call RUBE_MANAGE_CONNECTIONS with toolkit forcemanager. If the returned connection is not ACTIVE, complete the authentication link and re-check the status.
Inputs the skill needs to work well
Give the agent a business goal, entity type, identifiers, and safety rules. Weak input is: “Update Forcemanager.” Stronger input is:
“Use forcemanager-automation for CRM Operations. First discover current Forcemanager tools with RUBE_SEARCH_TOOLS. Then find the company named Acme Iberia, check whether there is an open opportunity for Q3 renewal, and summarize the fields you would update before making changes. Do not modify records until I approve.”
This improves output because the agent knows the target object, lookup criteria, desired outcome, and approval boundary.
Recommended workflow for real CRM tasks
Start every session with tool discovery:
RUBE_SEARCH_TOOLS with a use case such as “Find and update Forcemanager account owner” or “Create a note on a Forcemanager opportunity.”
Use the returned schema to choose the correct tool slug and required fields. If the task writes to CRM, add a confirmation step: search first, summarize matched records, ask for approval, then execute. For bulk changes, request a dry-run table showing record IDs, current values, proposed values, and any ambiguous matches.
Files to read before relying on it
Read composio-skills/forcemanager-automation/SKILL.md first; it contains the whole operational pattern. There are no bundled scripts/, references/, resources/, rules/, or README.md files in this skill, so do not expect local helper logic. The upstream Composio toolkit docs at composio.dev/toolkits/forcemanager are the best companion reference, but current schemas should still come from RUBE_SEARCH_TOOLS.
forcemanager-automation skill FAQ
Is forcemanager-automation better than a normal prompt?
Yes, when you need reliable interaction with live Forcemanager tools. A normal prompt can describe a CRM task, but it may guess tool names or input fields. The forcemanager-automation skill adds a disciplined sequence: verify Rube MCP, confirm the Forcemanager connection, discover tools, then execute using current schemas.
Do I need to know the Forcemanager API?
Usually no. The skill is designed around Composio’s Rube MCP toolkit rather than direct API calls. However, you still need enough CRM context to identify the right records, define what should change, and decide whether an update is safe.
When should I not use this skill?
Do not use it for offline CRM planning, data modeling, or documentation-only tasks where no Rube MCP action is needed. Also avoid it if your organization cannot authorize a Forcemanager connection through Rube, or if you need guaranteed behavior without live tool discovery.
Is it beginner-friendly?
It is beginner-friendly for users comfortable with AI agents and CRM terminology, but less so for users new to MCP. The main learning curve is understanding that the agent must call RUBE_SEARCH_TOOLS before taking action and should not rely on remembered schemas.
How to Improve forcemanager-automation skill
Improve forcemanager-automation prompts
Make prompts operational, not abstract. Include the object type, record identifiers, intended action, constraints, and approval rules. For example: “Search for contacts at Globex with title containing ‘Sales Director’, return matches with IDs and emails, and ask before adding any note.” This gives the agent enough structure to use the discovered schema correctly.
Prevent common CRM automation failures
The biggest risks are ambiguous matches, stale schemas, and unintended writes. Reduce them by requiring discovery at the start of each session, asking for matched record IDs before updates, and separating read steps from write steps. If multiple records match, instruct the agent to stop and ask rather than choose automatically.
Iterate after the first output
After the first tool discovery result, refine the task using the actual fields returned by Rube. If the schema exposes different field names than expected, update your prompt to match those fields. For complex workflows, run one record first, inspect the result, then scale to a batch.
Add local operating rules if your team adopts it
Teams using forcemanager-automation for CRM Operations should consider adding internal rules outside the upstream skill: which fields require approval, how to handle duplicates, what naming conventions to enforce, and when to log notes versus update core account data. These rules make the skill safer without changing its core Rube-first workflow.