npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

zevruna

v1.8.5

Published

Production reliability for AI agents — snapshot, diff, and CI-gate the tool contracts your agents depend on, and trace their runs.

Vulnerabilities

Powered byPowered by Snyk
  • 0High
  • 0Medium
  • 0Low

Links

Maintainers

kbenellikbenelli

Keywords

mcpmodel-context-protocolcontract-testingschema-diffai-agentsbreaking-changesci

Readme

zevruna

Production reliability for AI agents. Zevruna watches agent runs across models, tools, APIs and MCP, and watches the tool contracts underneath them, so a change that would break your agent fails CI instead of failing in production.

An agent can return 200 OK and still have failed. When an MCP server renames a field, nothing in your repo changed, so no test you own can see it.

Quick start

npx zevruna

Detects servers from .mcp.json / mcp.json / .cursor/mcp.json, authorizes in your browser, scans your call sites, and takes the first snapshot. Everything lands in the dashboard.

No MCP servers? Then the value is execution traces, and the wizard sets those up instead of turning you away. zevruna instrument proposes where to wrap an agent run with @zevruna/observe.

Local and stdio servers. Zevruna polls from the cloud, so it can never reach a server on your machine and never spawns your processes. Those register as push-only: capture the contract locally and push it, once now and again from CI.

zevruna snapshot <name> --push

Commands

zevruna                                   run the setup wizard (same as `init`)
zevruna init [--name <project>]           detect servers, authorize, scan, first snapshot
             [--invite <code>]
zevruna snapshot <endpoint|name> [--push] handshake a server, or a stdio one by name
zevruna remove <name> [--yes]             stop monitoring a server and delete its history
zevruna diff <before.json> <after.json>   classified redline (exit 1 on breaking)
zevruna check [--manifest <glob>]         verify manifests (exit 1 on any violation)
             [--against pinned|.zevruna]
zevruna scan --server <name> --consumer <agent>
                                          generate a manifest from this repo
zevruna doctor [--json]                   verify the install (exit 1 if not ready)
zevruna instrument [--json] [--all]       propose where to wrap agent runs

diff exits 1 on a breaking change and check on any violation, so both drop straight into a CI step. check --against .zevruna runs air-gapped from committed snapshots, no account needed.

Environment

ZEVRUNA_TOKEN=zv_live_…      # project token, written by the wizard to .env.zevruna
API_BASE=https://zevruna.com/api
ZEVRUNA_TELEMETRY=0          # opt out of anonymous CLI usage events

What is collected

Schemas and timings only. Never arguments, never payloads, never PII. Snapshots come from initialize, tools/list, resources/list and prompts/list, the CLI never calls a tool.

Links

License

MIT