Operate a persistent local terminal agent with gptme
Run gptme as a terminal-first autonomous agent that can edit files, execute commands, browse the web, and keep local project context.
npx skills add agentskillexchange/skills --skill operate-a-persistent-local-terminal-agent-with-gptme
Use gptme when a user wants a persistent local agent for coding or operational tasks from the terminal instead of a one-off chat session. The agent can work with local files, shell commands, browser access, vision, and computer-use style actions while keeping project-level context. The scope is a local terminal agent workflow for repeatable repo and system tasks, not a generic Python package or broad assistant product card.
What this skill actually does
Inputs and prerequisites: Python, local shell, configured LLM provider, optional browser/computer-use dependencies.
Setup notes: Install from the documented package or download path in the gptme docs, configure the desired model provider credentials, then run gptme from the project directory for terminal-first agent work.
Source and verification boundary: use https://gptme.org/docs/ as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Custom Agents workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.