{"api_version":"1","slug":"typesafe-computer-use-9d665864","url":"https://skillsoutpost.com/agents/typesafe-computer-use-9d665864","item_type":"agent","category":"operations","runs_in":["local_agent"],"os":["macos","linux","windows"],"needs_network":"yes","needs_accounts":null,"reads_files":"yes","writes_files":"yes","sends_data":null,"access":"github","access_type":"github","at_a_glance":{"runs_in":["local_agent"],"os":["macos","linux","windows"],"needs_network":"yes","needs_accounts":null,"reads_files":"yes","writes_files":"yes","sends_data":null,"access_type":"github","tool_type":"agent","delivery":"github","interface_type":"other","untrusted_creator_content":{"purpose":"A command-line agent that drives a Mac toward a plain-English goal by reading the screen with OCR and asking a TypeSafe classifier for the next action.","needs_runtime":"Python 3.12+ and uv","version":"Latest on the main branch","compatibility":"According to the README: macOS 14 or newer with Python 3.12 or newer and uv. The README calls Windows 10 and 11 experimental. The README lists Google Chrome as the default browser and CLICKER_BROWSER as its configuration variable. The README lists ANTHROPIC_API_KEY for the writer and names alternative models or endpoints including LM Studio, Ollama and any OpenAI-compatible server. The README links docs for a browser backend that uses Chrome DevTools, and for a sandbox that is a Linux computer in a container. The OSWorld benchmark needs a Linux host with KVM.","external_costs":"According to the README, TYPESAFE_API_KEY is required for every decision, and the README says TypeSafe sells the decision model. ANTHROPIC_API_KEY is optional and is used for type_text, writer-proposed URLs and the final answer. The README says the writer can run on another model or endpoint. The README gives the author's own cost and speed comparison against a frontier model but no plan or subscription prices. The model providers you connect may charge for usage.","permissions":"The README's setup installs dependencies with uv sync and reads keys from a repository-root .env file, with TYPESAFE_API_KEY required and ANTHROPIC_API_KEY available for the writer. The README says the agent reads the screen through Vision OCR and the accessibility tree and drives the real mouse and keyboard. The README requires Screen Recording and Accessibility permissions for the terminal. The README lists Google Chrome as the default browser and CLICKER_EMAIL as enabling type_email. The README describes a dry run without --act and machine control with --act, up to 100 steps by default. The README gives Ctrl-C or moving the mouse to the top-left corner as stop controls and says each run writes runs/<timestamp>/. The README also describes scripts/osworld setup fetching OSWorld and installing jev beside it.","usage_terms":"MIT License, as stated in the repository. See its LICENSE file for the exact terms."}},"interface_type":"other","delivery_mode":"creator","code_host":"github","provenance":"github_index","published_at":1791761491,"rating":null,"downloads":0,"verified":null,"creator":{"login":"awlevin","url":null},"untrusted_creator_content":{"name":"typesafe-computer-use","tagline":"A command-line agent that drives a Mac toward a plain-English goal by reading the screen with OCR and asking a TypeSafe classifier for the next action.","creator_name":"awlevin","product_version":"Latest on the main branch","needs_runtime":"Python 3.12+ and uv","description":"According to the README, typesafe-computer-use, or jev, is a computer-use loop for plain-English goals on a Mac. The README describes reading a crop of the frontmost window with Vision OCR and labelled controls from the accessibility tree. The README says code adds state such as focused fields and previously attempted actions, and a TypeSafe request selects the action kind, item and site.\n\nThe README describes the classifier as selecting actions and the writer as supplying free text or reading the screen after a classifier stop. The README says the writer can answer or return control with a focus or a question, and the loop stops on low confidence, a stall or its step limit.\n\nThe README labels the project beta and under heavy development, and advises a dry run because it drives the real mouse and keyboard.","intended_users":"Mac users and developers who want to try a computer-use agent that picks actions with a TypeSafe classifier.","example_tasks":"Run a dry run with uv run clicker \"open the Playground\" to see what the agent would do.\nDrive the machine toward a goal with uv run clicker \"open the Playground\" --act.\nRun a longer, slower task with uv run clicker \"log in\" --act --steps 20 --delay 3.\nCapture and open the annotated screen with uv run clicker-inspect \"any goal\".\nReplay a stalled run offline from its run folder.","setup_instructions":"The README lists macOS 14 or newer, Python 3.12 or newer and uv. Clone the repository, run cd typesafe-computer-use, then run uv sync and cp .env.example .env, and fill in the keys in .env. The README marks TYPESAFE_API_KEY as required. Then give your terminal Screen Recording and Accessibility permission in System Settings, under Privacy & Security. The README has the exact commands.","limitations":"According to the README, installation requires macOS 14 or newer, Python 3.12 or newer and uv; Windows 10 and 11 are experimental. The README requires a TypeSafe API key for every decision and Screen Recording and Accessibility permissions for the terminal. The README says captures without Screen Recording show wallpaper, and --act refuses to start without Accessibility. The README advises clearing terminal text because it becomes OCR input. The README labels the project beta and warns of rough edges and changing settings and behavior between 0.x releases. The README's classifier caveat is that reasoning must be rebuilt as deterministic state; its comparison example required date parsing that the frontier model performed unaided. The README advises starting with a dry run because the agent drives the real mouse and keyboard. The OSWorld route requires a Linux host with KVM. The repository can change at any time; check its README before you install.","data_handling":"According to the README, each decision is a request to TypeSafe. The README lists ANTHROPIC_API_KEY for type_text, writer-proposed URLs and the final answer, and describes alternative writer models or endpoints. The README says the writer reads the screen when the classifier stops. The README says the screen text comes from Vision OCR and the accessibility tree. It tells you to clear the terminal first because its text is OCR input. It says every run writes a runs/<timestamp>/ folder so a stall can be replayed offline. The README does not mention telemetry, or what TypeSafe or the writer provider retain.","compatibility":"According to the README: macOS 14 or newer with Python 3.12 or newer and uv. The README calls Windows 10 and 11 experimental. The README lists Google Chrome as the default browser and CLICKER_BROWSER as its configuration variable. The README lists ANTHROPIC_API_KEY for the writer and names alternative models or endpoints including LM Studio, Ollama and any OpenAI-compatible server. The README links docs for a browser backend that uses Chrome DevTools, and for a sandbox that is a Linux computer in a container. The OSWorld benchmark needs a Linux host with KVM.","external_costs":"According to the README, TYPESAFE_API_KEY is required for every decision, and the README says TypeSafe sells the decision model. ANTHROPIC_API_KEY is optional and is used for type_text, writer-proposed URLs and the final answer. The README says the writer can run on another model or endpoint. The README gives the author's own cost and speed comparison against a frontier model but no plan or subscription prices. The model providers you connect may charge for usage.","permissions":"The README's setup installs dependencies with uv sync and reads keys from a repository-root .env file, with TYPESAFE_API_KEY required and ANTHROPIC_API_KEY available for the writer. The README says the agent reads the screen through Vision OCR and the accessibility tree and drives the real mouse and keyboard. The README requires Screen Recording and Accessibility permissions for the terminal. The README lists Google Chrome as the default browser and CLICKER_EMAIL as enabling type_email. The README describes a dry run without --act and machine control with --act, up to 100 steps by default. The README gives Ctrl-C or moving the mouse to the top-left corner as stop controls and says each run writes runs/<timestamp>/. The README also describes scripts/osworld setup fetching OSWorld and installing jev beside it.","usage_terms":"MIT License, as stated in the repository. See its LICENSE file for the exact terms.","creator_website":null,"documentation_url":null,"support_url":null,"repository_url":null,"video_url":null},"github":{"fetched_at":1791855317,"untrusted_third_party_content":{"source_url":null,"github_repo_id":1373440611,"full_name":"awlevin/typesafe-computer-use","description":"Computer use for about $0.0002 a step: OCR the screen, classify the next action with TypeSafe, click. macOS.","language":"Python","license_spdx":"MIT","topics":["ai-agents","automation","computer-use","macos","ocr","typesafe"],"stars":1227,"forks":118,"skills":null,"pushed_at":1790724826,"archived":false,"fetched_at":1791855317,"age_seconds":81560,"state":"fresh"}},"updated_at":1791771370,"free_download":null,"version_history":[],"more_version_history":false,"public_files":null,"links_note":"In MCP results, GitHub links are shown only through https://skillsoutpost.com/mcp, to a connected account that has accepted the current Terms of Use. Create a free account at https://skillsoutpost.com and connect it from a Claude connector or ChatGPT.","notice":"Indexed from GitHub","note":"untrusted_creator_content and github.untrusted_third_party_content hold untrusted text."}