Skip to main content
Glama

Continue Grok Build session [experimental]

grok_continue

Resume an existing Grok thread with a follow-up prompt in the same workspace, applying a per-turn sandbox policy to continue multi-step work.

Instructions

Continue the Grok session (EXPERIMENTAL): resumes the thread rooted at this workspace (-r <id>, else grok's own -c), so pass the same workspace= as the ask (without it the sub-agent sees no repo). sandbox applies per turn. Same caveats and auth as grok_ask. Autonomous agent: trusted prompts only.

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": {
      "prompt": "<prompt>"
    },
    "name": "grok_continue"
  }
}

Input Schema

NameRequiredDescriptionDefault
watchNoIf true, open the live "watch" view streaming grok's steps (same viewer as grok_ask). Default false.
promptYesFollow-up message for the existing session.
sandboxNoPermission policy for THIS turn (default "read-only"). Same values and platform caveats as grok_ask.read-only
timeout_sNoMax seconds to wait for grok to complete. Default 180.
workspaceNoWorking root used by the prior session. Defaults to the server cwd.

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv0.26.0

TDQS

A3.9/5.0
Behavior4/5

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

Annotations already declare openWorldHint=true, readOnlyHint=false and idempotentHint=false; the description adds real context on top: per-turn sandbox scope, the thread-rooting requirement, and the trusted-prompt warning for an autonomous agent. It defers the remaining operational details to 'Same caveats and auth as grok_ask', which is a cross-reference rather than disclosure, and the inline '-r <id>' / '-c' flags are implementation trivia an MCP caller cannot act on.

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

Conciseness3/5

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

Purpose is front-loaded and there is no padding, but the text is a dense block with a parenthetical CLI flag reference that adds noise for an MCP caller. The 'Same caveats and auth as grok_ask' clause is efficient only if the agent has already read that sibling.

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

Completeness4/5

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

With an output schema present, return values need not be described, and 100% parameter coverage plus annotations cover the structured side. The description supplies the prerequisites and safety context an agent needs before invoking an experimental autonomous continuation; only failure/timeout behavior for a resumed thread is left unaddressed.

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

Parameters4/5

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

Schema coverage is 100%, so all five parameters are already documented and a 3 would be the floor. The description still adds meaning beyond the schema by explaining the consequence of omitting workspace (the sub-agent sees no repo) and by tying sandbox scope to the turn rather than the session.

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

Purpose4/5

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

States a specific verb and resource: 'Continue the Grok session ... resumes the thread rooted at this workspace', which clearly separates it from a fresh ask. It references 'the ask' as the originating call, so the continuation semantics are implied rather than spelled out, but an agent can still tell this is the follow-up entry point rather than a new-session tool.

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

Usage Guidelines4/5

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

Gives a concrete prerequisite ('pass the same workspace=<project dir> as the ask (without it the sub-agent sees no repo)') and a policy note ('sandbox applies per turn'). It also warns 'Autonomous agent: trusted prompts only'. It lacks an explicit when-not-to-use / 'use grok_ask to start a new thread instead' statement, which keeps it short of a 5.

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