Jules Dispatch

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/yuuqq-jules-dispatch

About

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list, install config and code. Checked Oct 25, 2026.

Tools

5 read · 4 write · 1 delete

Secrets

1 (API keys, tokens or passwords)

Install scripts

None found

Runs as

Node.js package

Tools (10)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

• jules_approve_plan

  Approve the current plan for a Jules session that is awaiting plan approval. Use this only after an AI agent has reviewed the latest plan and decided the session should continue executing. Returns: { success: true, data: { sessionId } } See also: jules_interact (review the latest plan before approving), jules_monitor

• jules_auto

  Plan a high-level intent with an OpenAI-compatible LLM and dispatch the resulting Jules tasks in one call. Use this optional one-shot fan-out tool when an AI agent is ready to create sessions immediately instead of reviewing planned tasks first. Dispatch uses the same bounded worker pool and optional global launch pacing as jules_dispatch. Returns: { success: true, data: { plan: { model, rationale?, tasks, usage? }, summary: { total, dispatched, failed }, results: [{ taskFile, taskTitle, sessionId, sessionUrl, title, status, error? }] } } See also: jules_plan_tasks (plan without dispatch), …

• jules_cancel_sessioncan change or delete data

  Cancel a running Jules session. Use this when an AI agent determines a session is no longer wanted, is blocked beyond recovery, or should be stopped before it makes further changes. Returns: { success: true, data: { sessionId } } See also: jules_monitor (check before cancelling), jules_dispatch (start replacement work)

• jules_dispatch

  Dispatch one or more Jules tasks through a single consolidated orchestration call. Use this when an AI agent has a single task object, an array of task objects, or a YAML/JSON task payload and wants a continuously replenished worker pool with bounded concurrency and optional global launch pacing. Returns: { success: true, data: { summary: { total, dispatched, failed }, results: [{ taskFile, taskTitle, sessionId, sessionUrl, title, status, error? }] } } See also: jules_monitor (track dispatched sessions), jules_interact (inspect one session in context), jules_list_sources (find source identi…

• jules_interactread-only

  Fetch full Jules session interaction context in one call. Use this when an AI agent needs the session record, derived status, globally latest plan, newest chronological activity window, total activity count, and PR output together before deciding whether to approve, message, monitor, or collect results. Returns: { success: true, data: { session: { id, title, state, url, source, branch }, status, plan, activityTotal, activities: [{ id, time, originator, title?, agentMessaged?, userMessaged?, planGenerated?, completed?, failed?, sessionFailed? }], pr? } } See also: jules_dispatch (create sess…

• jules_list_sessionsread-only

  List recent Jules sessions with optional pagination. Use this when an AI agent needs to discover recent sessions before selecting IDs for inspection or status checks. Returns: { success: true, data: { sessions: [{ name, id, title, url, state?, createTime? }], nextPageToken? } } See also: jules_monitor (status for known IDs), jules_interact (full session record)

• jules_list_sourcesread-only

  List all GitHub repositories connected to Jules. Use this to discover available source identifiers before dispatching or planning Jules work. Returns: { success: true, data: { sources: [{ id, name, githubRepo? }] } } See also: jules_dispatch (uses source identifiers), jules_list_sessions, jules_plan_tasks

• jules_monitorread-only

  Monitor one or more Jules sessions with optional waiting until terminal or action-required state. Sessions that cannot be polled at all (e.g. unknown IDs) are reported in the errored bucket while the rest keep polling. Use this for consolidated status checks, or set wait=true after dispatching when an AI agent needs completed/failed/cancelled/action-required/errored/still-running buckets before the next orchestration step. Returns without wait: { success: true, data: { sessions: [{ sessionId, title, state, status, prUrl?, lastActivity? }] } } Returns with wait: { success: true, data: { sess…

• jules_plan_tasksread-only

  Expand a high-level intent into independent, parallelizable Jules task definitions without dispatching them. Use this optional planner when an AI agent needs an LLM-assisted task breakdown to review or pass into jules_dispatch later. Returns: { success: true, data: { model, rationale?, tasks: [{ title, prompt, source?, branch?, autoMode?, requirePlanApproval? }], usage? } } See also: jules_dispatch (dispatch planned tasks), jules_auto, jules_list_sources

• jules_send_message

  Send a follow-up message to a running Jules session. Use this when an AI agent needs to unblock a session, answer a question, clarify requirements, or provide revised instructions. Returns: { success: true, data: { sessionId } } See also: jules_interact (inspect the conversation), jules_cancel_session

Signals

Stars

5

Forks

0

Open issues

0

Last commit

Oct 22, 2026 (8 days ago)

Last release

Aug 13, 2026 (2 months ago)

From GitHub (fetched Oct 24, 2026).

Details

Registry name

io.github.Yuuqq/jules-dispatch

Version

1.3.2

Auth

API key

Packages

npm @yuuqq/jules-dispatch

License

MIT

Repository

github.com/Yuuqq/jules-dispatch

First seen

Oct 24, 2026

Last checked

Oct 24, 2026

Source

Official MCP Registrygithub-graphql