Overview
Jevlin is a free web search for coding agents that asks twelve search providers at once and answers with one attributed list. Available in Claude Code, Cursor, Codex, opencode, Pi and Hermes.
Every search is observable on the platform. A trace shows one search with a receipt per provider: what each returned, how long it took, what it cost. A trajectory shows one agent session as the path it took, turn by turn: what the agent was working on when it searched, what came back, and what it concluded. Each search carries the context that makes this possible, hashed and scrubbed before it leaves the machine. See Trajectory observability.
Search is free for agents: setup registers you with the search platform and gets a key, with no card and nothing to top up.
Get started
Install it, then run setup. Setup registers your installation, writes a skill and hooks into each coding agent it finds, and prints a link to claim your agent. The quickstart walks through it.
- npm (recommended)
- Native install
- Manual download
npm install -g @jevlinsearch/jevlin
jevlin setup
Install it globally, never with npx. Setup writes the binary's path into your agents' skills and hooks, and npm throws away an npx cache or a project's node_modules. An npm install is updated with npm install -g @jevlinsearch/jevlin@latest.
macOS, Linux, WSL:
curl -fsSL https://jevlin.ai/install.sh | sh
Windows PowerShell:
irm https://jevlin.ai/install.ps1 | iex
Windows CMD:
curl -fsSL https://jevlin.ai/install.cmd -o install.cmd && install.cmd && del install.cmd
The installer downloads the latest release for your machine from the project's GitHub releases, checks the project's signature on that release and the archive's checksum, puts the binary at ~/.jevlin/bin/jevlin, and starts jevlin setup. The signature check uses openssl, node or python3 with cryptography, whichever is present; on Windows it uses node. Set JEVLIN_VERSION to install an exact release, or JEVLIN_NO_SETUP=1 to stop after the binary. A native install keeps itself current; see Upgrading.
Download the archive for your OS and architecture from the releases page and verify it against that release's checksums.txt. Put jevlin in a directory only you can write, such as ~/.local/bin, and run jevlin setup from there.
tar -xzf jevlin_<version>_<os>_<arch>.tar.gz
mkdir -p ~/.local/bin && mv jevlin ~/.local/bin/
~/.local/bin/jevlin setup
Setup refuses a binary under the temp directory, or in a directory another user owns or every user can write, because your agents' hooks and your PATH will name that path.
What you can do
- Search from your agent. The skill tells the agent to prefer this search over its built-in one and how to read the answer.
/jevlin <question>runs one search; naming jevlin in an ordinary request does the same. See Search from your agent. - Pick how wide a search goes.
fastasks one provider,balancedasks the core providers with each result attributed,maxasks every one for corroboration. You set the band, the agent asks for more when it needs it. See Providers and tiers. - See the whole trajectory. Each search carries hashed session, turn and call ids and the assistant text that led to it, so the platform shows a session as the path the agent took: what it asked, what came back, and what it concluded. See Trajectory observability.
- Keep control of what leaves the machine. Queries travel on stdin, never on the command line. The trace is scrubbed of secrets before it goes anywhere, and
JEVLIN_TRACE=offsends none. See What leaves the machine.
Use Jevlin everywhere
Any other agent gets a rules block to paste. See Any other agent.