{
  "edition": {
    "edition": "2026-09-06-draft-01",
    "language": "en",
    "publicationState": "published",
    "sourceRevision": "a7e788eb4b151fd92c1f58153e015773e3a829af",
    "projectCount": 67,
    "documentCount": 94,
    "skillCount": 10,
    "archivePath": "library.zip",
    "archiveBytes": 181812,
    "releasedAt": "2026-09-08T05:03:11.129+00:00"
  },
  "access": {
    "released": true,
    "rehearsal": false
  },
  "project": "background-services/batch-work-queue.md",
  "markdown": "https://legion.tw/api/projects/packet.md/background-services/batch-work-queue.md?edition=2026-09-06-draft-01",
  "documents": [
    {
      "path": "background-services/README.md",
      "text": "# Background automation and services\n\n[All project types](../README.md) · [Shared launch](../shared/launch.md)\n\nGive recurring work a dependable trigger, an understandable result, and a way for the person to inspect or control it.\n\n## Projects to build\n\n- [Watched-folder preparation](watched-folder.md)\n- [Recoverable backups](recoverable-backups.md)\n- [Meaningful change monitoring](change-monitor.md)\n- [Prepare tomorrow's information](tomorrow-briefing.md)\n- [Event-triggered follow-up administration](event-follow-up.md)\n- [Give the computer a batch of work](batch-work-queue.md)\n- [A personal overnight researcher](overnight-researcher.md)\n\nThese are instructions for projects you can build with your coding agent.\n"
    },
    {
      "path": "background-services/batch-work-queue.md",
      "text": "# Give the computer a batch of work\n\nBuild a persistent work queue for a processing routine the person wants to run across several items. Build it extremely well so they can submit work, do something else, and return to usable results. Follow the [shared launch](../shared/launch.md).\n\nChoose a real workload: recordings to transcribe, videos to convert, artwork to export, or documents to prepare. Use an existing application or processor where the person already has a working routine. Make that routine complete before judging the queue around it.\n\n## Let the person submit a clear workload\n\nGive them a convenient way to select the material, choose the processing settings, and see what will be queued. Keep each job associated with its intended source, settings, and output destination.\n\nMake the queue fit the task. The person may use one preparation setup for the whole batch and change it for a particular item. They should be able to recognize the items and correct a choice before it runs.\n\nPreserve original material and any useful existing output. Make a deliberate rerun or new output version understandable in the person's workflow.\n\n## Show the work as it happens\n\nDistinguish waiting jobs, active processing, completed output, and items needing attention. Show useful progress from the actual processor and make finished results available as soon as they are ready.\n\nLet the person inspect an output while the remaining batch continues. A failed item should retain its source and explanation while other independent work can proceed. Make retrying that item straightforward after the cause has been addressed.\n\nGive the person control over the remaining order, with a clear effect when they move, remove, or pause a job. Explain what pausing or cancelling means for the work currently running. Retain completed work and identify any partial output accurately.\n\n## Keep the workload through interruptions\n\nSave the queue and its results so an application restart or interrupted computer session leaves the work recoverable. Recognize completed jobs when resuming and make the remaining work clear.\n\nRun the processor in the appropriate place for the person's setup. Explain what needs to stay available and choose resource use that lets the computer serve their other work as intended. Make any model, service, or processing charges understandable before the batch runs.\n\nProvide a convenient view and controls through the existing application or another suitable local interface. The person should be able to inspect the actual worker, pause the queue, resume it, or stop the workload.\n\n## Return to finished work\n\nDeliver the working queue around the chosen processing routine. Submit representative items, use an early result while later work continues, rearrange something waiting, and recover from an interruption or failed item.\n\nJudge the actual outputs in the applications that use them. The queue should tell the person what finished, where it is, and what still needs attention. Apply the shared launch's ownership and recovery guidance to the setup, job records, and valuable results.\n\nExplain how the computer can hold a workload independently of the person's immediate attention. They can decide what to do next without personally starting every operation.\n\nLater, the [remote-agent phone app](../native-phone/coding-agent-in-your-pocket.md) or another companion can submit work and show completion away from the computer. Keep that connection attached to the same queue and results, so remote control extends the workload already here.\n"
    },
    {
      "path": "native-phone/coding-agent-in-your-pocket.md",
      "text": "# Talk to your coding agent from your phone\n\nBuild a native phone app through which the person can continue working with you on their computer while they are away. Build it extremely well, with a dependable conversation and useful control over real work happening in their projects.\n\nRead this brief and follow the [shared launch](../shared/launch.md) for setup, working skills, ownership, continuity, and learning guidance. Discover the build and installation route for their actual phone and carry it through to an installed application.\n\n## The experience\n\nThe person has a computer where their coding agent can work on applications. From their phone, they choose a project, open or start a conversation, and ask for something. The agent can inspect the project and carry out the work using its normal capabilities. They can read its response, discuss a decision, answer a question, and interrupt the work when their direction changes.\n\nImagine they notice an awkward screen in an app while out. They ask the agent to improve it, explain what bothered them, and then put the phone away. The computer continues working. When they return, they can see what happened and continue the same conversation.\n\nConnect the phone app to the coding-agent environment the person already uses, with its existing account and model access. Carry the conversations, working context, and actions between the phone and that environment. The person should recognize the project and agent they work with at their desk.\n\nRemote work should use the same project instructions, installed skills, and working knowledge. Use the installed agent's supported integration capabilities so the phone participates in real agent sessions and can accurately show what happened.\n\n## Stay oriented while away\n\nMake it clear which project and conversation are open. Keep messages readable on the phone, with useful progress and results easy to follow. The person needs to understand whether the agent is working, waiting for an answer, finished, or currently unreachable.\n\nGive their own actions equally clear outcomes: they can see whether a message has reached the computer and whether an interruption took effect. A lost phone connection and stopped agent work are different events; the experience should make that distinction understandable.\n\nThe conversation survives switching applications, locking the phone, and reconnecting later. When the person returns, show the current conversation and the work completed during their absence. Keep the underlying project and its work available from the computer as well.\n\n## Connect their own devices\n\nUse a private connection suited to the person's devices. Tailscale is a natural fit for reaching their computer while away. Establish the connection as part of making the application useful, with access belonging to the person and the devices they choose.\n\nExplain what needs to stay running on the computer for the agent to be available. Help the person understand any account, model-usage, or service costs involved in their chosen setup. Select the installation and connection approach from their actual environment; own those engineering decisions.\n\n## Build a foundation they can keep using\n\nThe complete experience here is remote conversation, real project work, direction, interruption, and returning to what happened. Give those capabilities sustained attention through focused steps, then put them together in an installed app the person can use from their phone.\n\nExercise the remote working experience, including leaving and returning to a conversation. Help the person try a useful change to one of their projects and understand where the result lives. Keep the interface centered on the work they are directing.\n\nInclude a connection from outside the computer's local network in that experience. Help the person understand the roles of their devices: the phone carries their conversation and control, while the computer runs the agent and does the project work. They should know how to restore the connection and continue when a device has been unavailable.\n\nLater, they may want voice conversations, photographs, completion notifications, previews, or a convenient way to install something the agent built. Notes captured during the day can become material for a conversation. Choose a foundation that allows those additions while preserving the conversations and projects they already rely on.\n"
    }
  ]
}