token-metrics-automation
by ComposioHQtoken-metrics-automation helps Claude run Token Metrics workflows through Composio Rube MCP with a schema-first pattern: verify Rube MCP, confirm an active token_metrics connection, search tools with RUBE_SEARCH_TOOLS, then execute the right workflow.
This skill scores 66/100, which means it is acceptable for directory listing but should be presented as a lightweight MCP workflow wrapper rather than a fully worked Token Metrics automation playbook. Directory users get enough clarity to understand when to install it and how an agent should start, but execution still depends heavily on live Rube tool discovery and external toolkit schemas.
- Clear trigger and scope: it is specifically for automating Token Metrics operations through Composio's Token Metrics toolkit via Rube MCP.
- Includes essential prerequisites and setup checks, including requiring Rube MCP, `RUBE_SEARCH_TOOLS`, and an active `token_metrics` connection via `RUBE_MANAGE_CONNECTIONS`.
- Provides an agent-safe discovery-first pattern that tells the agent to fetch current tool schemas before execution, reducing risk from stale tool assumptions.
- Operational detail is mostly delegated to `RUBE_SEARCH_TOOLS`; the evidence does not show concrete Token Metrics task examples, schemas, or end-to-end workflows beyond the discovery/connection pattern.
- No supporting scripts, references, resources, README, or install command are present, so adopters must rely on the single SKILL.md and external Composio/Rube behavior.
Overview of token-metrics-automation skill
What token-metrics-automation does
token-metrics-automation is a Claude skill for running Token Metrics workflows through Composio’s Rube MCP server. Its main purpose is not to hard-code one Token Metrics action, but to make the agent discover the current Token Metrics tool schemas first, confirm the user’s Token Metrics connection, and then execute the right MCP tool with fewer assumptions.
This matters because Token Metrics automation depends on live tool availability, authentication state, and changing input schemas. A generic prompt may invent fields or call the wrong tool; this skill pushes the agent toward RUBE_SEARCH_TOOLS before execution.
Best-fit users and workflows
The token-metrics-automation skill is best for users who already use Claude with MCP and want to automate crypto research or Token Metrics operations through Composio. It fits workflows such as finding available Token Metrics actions, preparing a structured tool call, checking connection status, and chaining Token Metrics data into a broader research or reporting workflow.
It is most useful for operators, analysts, and automation builders who want a repeatable “discover, validate, execute” pattern rather than one-off manual prompting.
Key differentiator: schema-first automation
The strongest differentiator is the skill’s instruction to always search tools first. Instead of assuming a fixed API shape, the agent should call RUBE_SEARCH_TOOLS with a specific Token Metrics use case, review returned tool slugs and schemas, then proceed.
That makes the skill safer for Workflow Automation where integrations change over time. The tradeoff is that the first step may feel slower than asking for a direct answer, but it reduces failed calls and hallucinated parameters.
What to check before installing
Before you use token-metrics-automation, confirm that your AI client supports MCP servers and that you can add https://rube.app/mcp as a server. The upstream skill has a single source file, SKILL.md, so installation decisions should focus on whether this schema-discovery pattern matches your workflow rather than expecting a large library of helper scripts.
How to Use token-metrics-automation skill
token-metrics-automation install context
A typical token-metrics-automation install is through an AI skill manager, for example: npx skills add ComposioHQ/awesome-claude-skills --skill token-metrics-automation.
After adding the skill, configure Rube MCP in your client with https://rube.app/mcp. Then verify that RUBE_SEARCH_TOOLS is available. Use RUBE_MANAGE_CONNECTIONS with toolkit token_metrics to check whether your Token Metrics connection is ACTIVE. If it is not active, complete the returned authorization flow before requesting any Token Metrics operation.
Inputs the skill needs from you
Give the agent a specific Token Metrics task, the intended output, and any constraints. Weak input is: “Use Token Metrics for Bitcoin.” Strong input is: “Using Token Metrics via Rube MCP, discover the available tools first, then retrieve the most relevant BTC token analytics available through the active token_metrics connection. Return a short analyst brief with the tool used, key fields returned, and any missing data.”
Useful details include token symbols or IDs, timeframe, report format, whether you need raw data or a summary, and whether the result will feed another automation step.
Practical token-metrics-automation usage pattern
A good token-metrics-automation usage flow is:
- Ask the agent to confirm Rube MCP availability.
- Ask it to run
RUBE_SEARCH_TOOLSfor the exact Token Metrics use case. - Review the returned tools, schemas, and pitfalls.
- Have the agent choose the best matching tool and explain required fields.
- Execute only after the Token Metrics connection is active.
- Ask for a concise result with tool names, parameters used, and limitations.
This sequence is especially important for token-metrics-automation for Workflow Automation because later workflow steps often depend on predictable fields and error handling.
Repository files to read first
Start with composio-skills/token-metrics-automation/SKILL.md. It contains the prerequisites, setup steps, tool discovery pattern, and the core workflow. There are no visible support folders such as scripts/, references/, or resources/ in the provided tree, so the skill’s operational value is concentrated in that single file.
When reviewing it, focus on the required MCP dependency, the Token Metrics connection requirement, and the repeated instruction to search current schemas before calling tools.
token-metrics-automation skill FAQ
Is token-metrics-automation beginner-friendly?
It is beginner-friendly if you are already comfortable adding an MCP server to Claude or another compatible client. It is not a general “explain crypto” skill. The main learning curve is understanding that the agent must discover tools through Rube MCP and authenticate the Token Metrics toolkit before useful automation can happen.
How is this better than an ordinary prompt?
An ordinary prompt may answer from general knowledge or guess an integration shape. The token-metrics-automation skill adds operational guardrails: check Rube MCP, manage the Token Metrics connection, search tools for current schemas, and only then execute. That makes it better for real tool use than for purely conversational crypto commentary.
When should I not use this skill?
Do not use it if you only need educational crypto explanations, if your client cannot connect to MCP servers, or if you do not have access to an active Token Metrics connection through Composio. It is also a poor fit when you need guaranteed deterministic production pipelines without human review, because live tool discovery and returned schemas can change.
Does it include scripts or custom code?
No. Based on the repository structure, token-metrics-automation is a prompt-and-workflow skill centered on SKILL.md, not a package with executable scripts. Its value is in instructing the agent how to use Rube MCP correctly.
How to Improve token-metrics-automation skill
Improve token-metrics-automation prompts
To get better results, make the prompt explicit about discovery, execution, and output shape. For example: “Use token-metrics-automation. First run RUBE_SEARCH_TOOLS for Token Metrics tools that can analyze ETH market signals. If multiple tools match, compare them briefly, choose the best one, execute only with the active token_metrics connection, then return a table of fields retrieved and a short interpretation.”
This gives the agent a task, a decision rule, and a reporting format.
Reduce common failure modes
The most common failure modes are skipping tool discovery, assuming stale parameters, attempting execution before authentication, and producing a summary without naming the tool used. Counter these by requiring the agent to show the discovered tool slug, required fields, connection status, and any missing inputs before it runs an action.
If the agent cannot find a suitable tool, ask it to report that clearly instead of improvising.
Iterate after the first output
After the first run, improve the workflow by asking: “Which fields were unavailable?”, “Which schema fields would improve precision?”, “Can this be narrowed by token, date range, or metric type?”, and “What should the next automation step consume?”
This turns token-metrics-automation from a one-shot request into a reusable Workflow Automation step with cleaner handoffs.
Strengthen the skill for team use
For shared use, document your preferred Token Metrics tasks, approved output formats, and review requirements in your own project notes. Add prompt snippets for common jobs such as token screening, metric extraction, or analyst briefs. The upstream skill is intentionally lean, so local conventions are what make token-metrics-automation more reliable across a team.