Taprun is a local-first tool designed to automate browser actions in a user's own authenticated Chrome session without transmitting credentials to external servers. It addresses the challenge of automating tasks behind login walls, such as OTP-gated pages, Cloudflare Turnstile checkpoints, and dashboards that require active sessions, by operating directly within the user's existing browser environment. This approach ensures that authentication tokens and cookies remain on the user's machine, never leaving local storage.
The platform operates by compiling a user's desired browser task into a deterministic JSON plan file using a one-time AI process. This plan can then be replayed infinitely in the local browser at zero additional token or cloud compute cost. Taprun's replay mechanism enables users to act, verify outcomes, and re-run tasks if web page structures drift over time. The verification feature uses snapshot equivalence to compare current web page states with baselines, providing clear verdicts on whether a site has changed, is equivalent, or is unreachable. If a site changes, users can re-capture the plan to adapt to the new structure.
Taprun is delivered as a Chrome extension that bridges the user's authenticated session to the automation process, as well as a command-line interface (CLI) installable via npm, brew, or curl. It supports Mac, Linux, and Windows platforms, with all distributions signed. The tool can also be connected to compatible MCP (Multi-Component Pipeline) hosts through standard input/output connections.
The service is open source and released under the MIT license. Its pricing model is based on a one-time cost at compile time for the AI planning step, with subsequent replays incurring no additional token or cloud charges. Community-contributed "taps"—prebuilt automation plans—are available, and users are not required to create accounts or provide API keys. Taprun positions itself as an alternative to cloud-based browser automation SDKs by ensuring privacy and local control over sensitive session data.
In the Frameworks & SDKs space, taprun takes a focused approach. It focuses on automating and replaying AI agent browser actions deterministically without incurring LLM token costs. taprun is an open-source project aimed at AI developers. The project is open source (MIT). It runs on the web, the command line, and API, and it can be self-hosted.
LeonTing1010 builds and maintains taprun, and the product first shipped in 2026. The project is developed in the open on GitHub with 563 commits in the last 90 days. Across PulseGate's embedding index, taprun has few near neighbours, marking it as relatively distinct. Among its 6 catalogued features are browser automation, deterministic replay, and Python SDK. It exposes integrations via an MCP server.
Latest indexed changes and source events
taprun.dev discovered by the PulseGate indexer
Other apps tracked under the same category.
