DEDUPED Hot AI repos, turned into use cases Get 5 a day

How to serve a developer seat from a local GPU model with Auto Swarm Shadow Adapter

Connect a local Ollama model to the senior_developer role using shadow-worker.

Credentials. The worker command takes a username and password on the command line, so use an account created for this node and avoid saving the password in shared shell history.

Who it is forAn engineer with a GPU workstation who wants AutoSwarm's senior_developer tasks handled by a local model instead of a cloud API.

The problemAutoSwarm tasks assigned to a developer seat need a worker to pick them up, and the engineer wants the work done locally on their own model.

The outcomeThe shadow-worker process logs in to the adapter server as a named user, claims senior_developer tasks, and runs them through the local Ollama model.

What you need first

  • A working AutoSwarm install (the README links to HKUDS/AutoSwarm)
  • Python
  • A SHADOW_JWT_SECRET of at least 32 characters
  • Access to a local model server (Ollama) or an API key for Anthropic or OpenAI, depending on the use case

Steps

  1. Start the Ollama server at http://localhost:11434 and run export OLLAMA_HOST="http://localhost:11434".
  2. Run shadow-worker --server-url http://192.168.1.100:8800 --username <username> --password <password> --role senior_developer --provider ollama --model llama3.3:70b with your own account details. (inferred, check the README)
  3. Leave the worker running and confirm it stays connected to the server.
  4. Start an AutoSwarm session with opc chat --mode company "Build feature X" so a senior_developer task is created and assigned.
  5. Check the portal or store for the worker's submitted deliverable and the task's new phase.

Prompt for your agent

Use the repo AhmadHassan-BTed/AutoSwarm-Shadow-Adapter to connect a shadow-worker for the senior_developer role at <server URL> using the Ollama provider with model llama3.3:70b. Expected output: a running worker that reports the tasks it claims and the deliverables it submits. Verify that the worker logged in with the account <username> and that a claimed task reached the submitted state.

Caveat. The README example uses the llama3.3:70b model, so the machine must have enough memory to run that model in Ollama.

Share: X LinkedIn Hacker News Email Copy link

Disclaimer

Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page