Skip to main content
Glama
Sign In

mFlow

by io.m-flow
OverviewDiscussionsAdmin

assign_to_coding_agent

Assign a standalone task to mFlow's autonomous coding agent for asynchronous execution. Returns immediately with the codeAgentTaskId and queued status (plus queuePosition while it waits for a free sandbox) — it does not stream or wait for completion; poll get_coding_agent_status separately. Use this once you've decided a task is ready for implementation and want to delegate the "how" rather than doing it yourself. The task must already exist (create_standalone_task first if it doesn't).

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "assign_to_coding_agent"
  }
}

Input Schema

NameRequiredDescriptionDefault
tierNoSandbox tier preset: light (2GB/1 CPU) for small/simple changes, mid (3GB/2 CPUs) for typical work, heavy (7GB/4 CPUs) for large builds or heavier dependency installs. Defaults to mid when omitted.
taskIdNoInternal standalone task ID (Mongo _id) to assign. Provide exactly one of taskId or taskKey — taskKey is looked up via get_standalone_task_by_key if given.
repoUrlNoGitHub repo to work against, as "owner/repo" or a full github.com URL. Omit to have the agent scaffold a fresh project instead of targeting an existing repo.
taskKeyNoHuman-readable standalone task key, e.g. "MVPB-120". Provide exactly one of taskId or taskKey.
aiProviderNoBYOK AI provider to run the task with; omit to use mFlow's hosted/default provider.
extraPromptNoAdditional instructions appended to the task's existing title/description — use to give the agent context the task itself doesn't capture.
instructionsNoAlias for extraPrompt — provide one or the other, not both.
noRepoOutputNoOnly used when repoUrl is omitted: where the finished work goes. "zip" (default) saves the whole workspace as a zip linked on the task and attached to the standalone task — no GitHub needed. "new_repo" creates a new private GitHub repo on the user's connected account and pushes the work there (fails to start if GitHub isn't connected). Ignored when repoUrl is set.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / noRepoOutput
      Added value: +{
+  "description": "Only used when repoUrl is omitted: where the finished work goes. \"zip\" (default) saves the whole workspace as a zip linked on the task and attached to the standalone task — no GitHub needed. \"new_repo\" creates a new private GitHub repo on the user's connected account and pushes the work there (fails to start if GitHub isn't connected). Ignored when repoUrl is set.",
+  "enum": [
+    "zip",
+    "new_repo"
+  ],
+  "type": "string"
+}
  2. Changed7 schema fields changed
    • changedInput schema / properties / aiProvider / description
      Previous value: -"BYOK AI provider; omit to use mFlow hosted/default"New value: +"BYOK AI provider to run the task with; omit to use mFlow's hosted/default provider."
    • changedInput schema / properties / extraPrompt / description
      Previous value: -"Additional instructions to append to the task description"New value: +"Additional instructions appended to the task's existing title/description — use to give the agent context the task itself doesn't capture."
    • changedInput schema / properties / instructions / description
      Previous value: -"Alias for extraPrompt"New value: +"Alias for extraPrompt — provide one or the other, not both."
    • changedInput schema / properties / repoUrl / description
      Previous value: -"GitHub repo (owner/repo or full github.com URL) if targeting an existing repo"New value: +"GitHub repo to work against, as \"owner/repo\" or a full github.com URL. Omit to have the agent scaffold a fresh project instead of targeting an existing repo."
    • changedInput schema / properties / taskId / description
      Previous value: -"Internal standalone task ID (Mongo _id)"New value: +"Internal standalone task ID (Mongo _id) to assign. Provide exactly one of taskId or taskKey — taskKey is looked up via get_standalone_task_by_key if given."
    • changedInput schema / properties / taskKey / description
      Previous value: -"Human-readable standalone task key, e.g. \"MVPB-120\""New value: +"Human-readable standalone task key, e.g. \"MVPB-120\". Provide exactly one of taskId or taskKey."
    • changedInput schema / properties / tier / description
      Previous value: -"Sandbox tier preset: light (2GB/1 CPU), mid (3GB/2 CPUs, default), heavy (7GB/4 CPUs). Defaults to mid when omitted."New value: +"Sandbox tier preset: light (2GB/1 CPU) for small/simple changes, mid (3GB/2 CPUs) for typical work, heavy (7GB/4 CPUs) for large builds or heavier dependency installs. Defaults to mid when omitted."
  3. First observed

TDQS

A4.7/5.0
Behavior5/5

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

Beyond the annotations (which only say this is a non-destructive write), the description discloses the async contract: it returns immediately with codeAgentTaskId plus queued status and queuePosition, does not stream or block, and requires separate polling. That is exactly the behavioral context an agent needs to avoid waiting on a call that returns instantly.

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?

Three sentences, front-loaded with the action and its async nature, then the usage condition, then the prerequisite. No restatement of the title and no filler; each sentence carries distinct information.

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 fire-and-forget dispatch tool with no output schema, the description fully covers what matters: immediate return shape (codeAgentTaskId, queued status, queuePosition), the absence of streaming, the polling route, and the task preconditions. Nothing an agent needs to invoke and then follow up is missing.

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 description coverage is 100% and every parameter already carries a detailed description (tier presets, repoUrl formats, noRepoOutput behavior), so the schema does the heavy lifting. The description adds only the task-existence precondition implied by taskId/taskKey and says nothing about tier or provider selection, so it stays at the baseline.

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?

The description opens with a specific verb, resource, and destination: assign a standalone task to mFlow's autonomous coding agent. It is immediately distinguishable from sibling tools such as create_standalone_task (task creation) and get_coding_agent_status (polling), so an agent can route correctly without opening a 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 states the exact decision point for use ('once you've decided a task is ready for implementation and want to delegate the "how"'), names the prerequisite ('the task must already exist — create_standalone_task first'), and names the follow-up alternative ('poll get_coding_agent_status separately'). When-to-use, precondition, and hand-off path are all explicit.

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

All mFlow toolsInspect
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.

Resources