timely-automation

Overview of timely-automation skill

What timely-automation does

timely-automation is a Claude skill for automating Timely operations through Composio’s Rube MCP server. Its main value is not a fixed Timely script; it teaches the agent to discover the current Timely tool schemas with RUBE_SEARCH_TOOLS, verify the Timely connection, then execute the right Rube MCP tool for the task.

Use it when you want an AI assistant to handle Timely workflow automation with less guesswork, especially for actions where tool names, required fields, or API schemas may change.

Best-fit users and jobs

The timely-automation skill fits users who already work in Timely and want Claude to help with operational tasks such as finding available Timely actions, preparing tool calls, checking connection status, and running supported workflows through Composio. It is most useful for teams that prefer tool-driven automation over manual time-tracking administration.

It is less useful if you only need natural-language advice about time tracking, do not use Timely, or cannot connect Rube MCP in your AI client.

Key differentiator: schema discovery first

The most important behavior in this timely-automation guide is the “search tools first” rule. Instead of assuming a static API contract, the skill requires RUBE_SEARCH_TOOLS before execution. That matters because Composio tool schemas can evolve, and a stale prompt can fail silently or send incomplete fields.

The skill’s practical workflow is: discover tools, verify the timely connection, inspect required inputs, execute, then review the result.

Adoption considerations

Before installing timely-automation, confirm that your client supports MCP and that you can add https://rube.app/mcp as a server. The source skill has a single SKILL.md and no helper scripts or local resources, so adoption depends on your MCP setup, Timely authorization, and prompt quality rather than repository-side utilities.

How to Use timely-automation skill

timely-automation install context

Install the skill from the GitHub skill repository with:

npx skills add ComposioHQ/awesome-claude-skills --skill timely-automation

Then configure Rube MCP in your AI client by adding:

https://rube.app/mcp

The source skill says no separate API key is needed for the MCP endpoint, but you still need an active Timely connection. In practice, ask the agent to check that RUBE_SEARCH_TOOLS is available before doing any Timely work.

Required setup before running workflows

A reliable timely-automation usage flow starts with connection validation:

1. Confirm RUBE_SEARCH_TOOLS responds.
2. Use RUBE_MANAGE_CONNECTIONS with toolkit timely.
3. If the connection is not ACTIVE, follow the returned authorization link.
4. Re-check status before asking the agent to execute any Timely action.

Do not skip this step. Many apparent “AI failures” are actually inactive app connections or missing schema fields.

Turn a rough goal into a usable prompt

Weak prompt:

Update my Timely stuff.

Stronger prompt:

Use the timely-automation skill. First call RUBE_SEARCH_TOOLS for the exact Timely task: “find supported operations for managing Timely entries and projects.” Check the timely connection with RUBE_MANAGE_CONNECTIONS. If active, show me the available tool options, required fields, and any risks before executing. Do not run a write action until I approve the prepared tool call.

This works better because it names the skill, forces schema discovery, separates read and write actions, and asks for a pre-execution review.

Files to read first

Start with composio-skills/timely-automation/SKILL.md. It contains the full operational contract: prerequisites, setup, tool discovery, and the core workflow pattern. There are no README.md, metadata.json, rules/, resources/, or scripts/ files in this skill path, so the main file is the source of truth.

For tool-specific behavior, use the linked Composio Timely toolkit documentation and live RUBE_SEARCH_TOOLS output rather than guessing from examples.

timely-automation skill FAQ

Is timely-automation for Workflow Automation or reporting?

The timely-automation skill is primarily for Workflow Automation through Timely’s Composio toolkit. It can help discover tools that may support reporting-like tasks, but its dependable scope is whatever RUBE_SEARCH_TOOLS returns for your connected Timely account at runtime.

How is this better than an ordinary prompt?

An ordinary prompt may invent Timely API fields or rely on outdated assumptions. timely-automation instructs the agent to query Rube MCP for current tool schemas first, which makes the workflow more grounded and safer for execution.

Can beginners use it?

Yes, if they can configure MCP and complete the Timely connection flow. Beginners should ask the agent to explain each discovered tool and show required fields before execution. If you are not comfortable approving tool calls, start with read-only discovery tasks.

When should I not use timely-automation?

Do not use it when the Timely connection is inactive, when your client cannot access Rube MCP, or when the task requires unsupported Timely behavior. Also avoid using it for vague write requests like “clean up my account” unless you provide clear constraints and require approval before changes.

How to Improve timely-automation skill

Improve timely-automation prompts with task detail

Better inputs produce better tool selection. Include:

• The exact Timely object you care about, such as time entries, projects, users, clients, or tags
• Whether the task is read-only or can modify data
• Date ranges, project names, user names, or filters
• Approval requirements before write actions
• Desired output format, such as summary, table, or prepared tool call

This reduces wasted discovery and helps the agent map your goal to the right Composio tool.

Common failure modes to prevent

The most common problems are inactive Timely authorization, skipped tool discovery, incomplete required fields, and overbroad write requests. Prevent them by requiring the agent to:

1. Search tools for the specific use case.
2. Verify the timely connection.
3. Present the selected tool slug and schema.
4. Ask for missing fields.
5. Request approval before destructive or large-scale changes.

Iterate after the first output

After the first discovery response, refine the workflow instead of restarting. For example:

Use the same Rube session. Narrow the Timely tool search to updating time entries for last week. Compare the required fields across candidate tools and recommend the safest one before execution.

This keeps context, reduces repeated setup, and improves decision quality.

Add local operating rules if needed

If your team uses Timely in a specific way, extend your local prompt or project instructions with rules such as “never modify locked periods,” “always filter by workspace,” or “summarize before executing bulk changes.” The upstream timely-automation skill is intentionally lightweight, so local guardrails are the best way to adapt it to your workflow without changing the core discovery-first pattern.