Marsell

Agent run early

agent_run_early

Ask one of the agents the month's plan runs for one extra run now, before its turn comes: worth it when fresh findings would change what Marsell makes or answers in the next days and its last run is too old for that (the day before a premiere or a launch, to read where viewers or buyers are talking about it; news in the business's field; a rival's move). Not worth it when nothing this week depends on it. It spends from that agent's own tokens for the month and stops at its budget. Only an agent that is on; while one early run of it waits or runs, a second ask is refused. On now: search (AI answer checks, once a day); discovery (Discovery, once a day); listening (Listening, every hour); creators (Creator search, once a week); press (Press and directories, once a day); watch (Site checks, every hour); videos (Video readings, once a day). Returns the agent started and the tokens it had left this month before this run; what it finds shows where its work always shows (agent_open). It needs an act key, and spends the owner's tokens as the chat's same tool would: say the price before you spend.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "agent": "<agent>",
      "reason": "<reason>"
    },
    "name": "agent_run_early"
  }
}

Input Schema

Name	Required	Description
`agent`	Yes	The agent, by its id from the list of those on now.
`reason`	Yes	One plain sentence the owner reads in the activity feed, in the business's terms ("Your premiere is tomorrow, so it reads where viewers are talking about it.").
`project`	No	The project's id or exact name. Optional when the account has one project.

Schema Changelog

Changes observed during successful MCP inspections.

First observedNov 4, 2026

TDQS

A4.9/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the annotations: it discloses token spend charged to that agent's monthly budget with a hard stop, that only an agent currently "on" qualifies, and a concurrency rule (a second ask is refused while one run waits or runs). It also states the return payload, the act-key/auth requirement, and that it spends the owner's tokens, adding the caution to state the price before spending.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The core action is front-loaded, then conditions, constraints, and returns follow in a logical order. The inline agent-cadence list is dense and the opening clause is clause-heavy, but nearly every sentence adds decision-relevant information rather than filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 3-parameter mutation-style tool with no output schema, the description covers cost, budget cap, eligibility, concurrency, auth, and the shape of the return and where results appear. Nothing an agent needs in order to call it correctly or warn about spend is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is already 100%, but the description adds real meaning: it maps each enum value to the agent's role and cadence (listening every hour, creators once a week, etc.), which tells the agent why one id fits better than another. It also frames `reason` as the owner-facing sentence in the activity feed and `project` as only needed with multiple projects.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource: trigger one extra run of one of the named month-plan agents ahead of its scheduled turn. It enumerates the eligible agents and points to agent_open for where results surface, so an agent can distinguish it from agents_read / agent_open without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit when-to-use (fresh findings would change what Marsell makes or answers in the next days, last run too old, before a premiere or launch, news in the field, a rival's move) and explicit when-not-to-use ("Not worth it when nothing this week depends on it"). It also routes elsewhere for reading results (agent_open), leaving no inference needed.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Marsell tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.