Skip to main content
Glama

Run workflow

run_workflow
Destructive

Start a run. Spends credits, so the user must approve the cost first.

Needs estimate_run's run_token and the SAME script_text. Without confirm=true it starts nothing and returns the price to show the user; call again with confirm=true once they agree. Returns run_id and status; then poll get_run_status.

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": {
      "workflow_id": "<workflow_id>",
      "run_token": "<run_token>"
    },
    "name": "run_workflow"
  }
}

Input Schema

NameRequiredDescriptionDefault
confirmNofalse returns the price and starts nothing. true SPENDS credits — only after the user agrees.
run_tokenYesFrom estimate_run, bound to that workflow, graph and script_text. Expires in 10 minutes and is spent by the first run that starts.
script_textNoReplaces the workflow's stored script. estimate_run and run_workflow must get the SAME text.
workflow_idYesThe workflow_id from list_workflows. Never invent one.
dictionary_modeNoWhich dictionary entries this run applies. Omit to use the workflow's setting.
source_languageNoLocale of script_text, e.g. `ko-kr`. Must match the one given to estimate_run.
dictionary_entry_idsNoEntry ids from scan_script_for_dictionary_entries. Only with dictionary_mode='selected'.

Output Schema

NameRequiredDescriptionDefault
run_idNo
statusNo
createdNo
next_stepNo
max_creditsNo
workflow_nameNo
workspace_slugNo
current_balanceNo
outcome_chargesNo
run_console_urlNo
expected_creditsNo
workflow_summaryNo
requires_confirmationNo
estimated_overage_centsNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedOutput schema / properties / outcome_charges
      Added value: +{
      +  "items": {
      +    "additionalProperties": true,
      +    "type": "object"
      +  },
      +  "type": [
      +    "array",
      +    "null"
      +  ]
      +}
  2. Changed2 schema fields changed
    • addedInput schema / properties / dictionary_entry_ids
      Added value: +{
      +  "anyOf": [
      +    {
      +      "items": {
      +        "type": "string"
      +      },
      +      "type": "array"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "description": "Entry ids from scan_script_for_dictionary_entries. Only with dictionary_mode='selected'.",
      +  "title": "Dictionary Entry Ids"
      +}
    • addedInput schema / properties / dictionary_mode
      Added value: +{
      +  "anyOf": [
      +    {
      +      "enum": [
      +        "all",
      +        "none",
      +        "selected"
      +      ],
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "description": "Which dictionary entries this run applies. Omit to use the workflow's setting.",
      +  "title": "Dictionary Mode"
      +}
  3. First observed

TDQS

A4.7/5.0
Behavior5/5

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

Annotations declare destructiveHint=true and non-idempotent but give no operational detail; the description adds that credits are spent, that user approval is a prerequisite, that the token expires in 10 minutes and is consumed by the first successful start, and that confirm=false is a true no-op. These are non-obvious failure modes an agent would otherwise trip over.

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

Conciseness5/5

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

Four short lines, front-loaded with the cost warning, then the token/script prerequisite, then the confirm gate, then the follow-up call. No filler and the riskiest fact appears first.

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 destructive, credit-spending, two-phase tool with 7 parameters, the description covers prerequisites, the approval gate, token lifetime, and the post-call polling step. Return values need not be detailed because an output schema exists.

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

Parameters3/5

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

Schema coverage is 100% and the schema already documents confirm, run_token, and script_text in similar language, so the description largely restates structured data. Its one genuine addition is stressing that estimate_run and run_workflow must receive identical script_text, which tightens the coupling between the two calls. Baseline 3 is appropriate.

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?

"Start a run" is a specific verb+resource, and the description immediately distinguishes this from estimate_run (pricing only) and get_run_status (polling after). An agent can place this tool in the pipeline 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?

It lays out the exact call sequence: estimate_run first to obtain run_token with the SAME script_text, then call with confirm=false to surface a price, then repeat with confirm=true after user approval, then poll get_run_status. Explicit when-to-use, when-not (no confirm = no run), and the named next step.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.