claude

jev

Puts a fast calibrated classifier at the harness boundaries: hands Claude a note about a risky tool call, denies the two block-grade cases once with a reason Claude can answer, flags tool results that read like prompt injection, and catches a turn that stops short of the work it was asked to do. Advisory to the agent and fail-open; a command that matches a hard rule (rm -rf ~, a force push to main, turning the plugin off) asks you instead.

27 on GitHubby Brainwiresv0.7.2MITupdated todaysecurity
/

Plugin jev from marketplace brainwires-jevwire in Brainwires/jevwire. Commands read from the repo's manifest — today.

This plugin runs code on your machine — its hooks fire automatically on events like tool use and it starts local MCP servers. The exact commands are listed below; read them before installing.

What it installs

Skills (1)

  • jev-decisions

    Use when a decision in this session is a bounded judgment over text that would otherwise cost a full model turn — ranking or filtering more than about fifteen candidates, checking whether a source actually supports each claim before reporting it, choosing a next step after a confusing tool failure,

Slash commands (6)

  • /jev:calibrate

    Report how the jev gate has actually behaved — what Claude was told, what was suppressed, how every tripwire ended, and how many notes a different threshold would have produced.

  • /jev:daemon

    Show, stop or restart jev's loopback judgment daemon — the process the http hooks post to. Use it when /jev:status says the daemon is down, after a plugin update, or to check what it has served.

  • /jev:off

    Turn the jev hooks off for this session — no notes, no tripwires, no result screening, no stop check — until /jev:on.

  • /jev:on

    Turn the jev hooks back on after /jev:off, clearing both the session flag and the global one.

  • /jev:status

    Show what the jev hooks are configured to do, and what they have done in the last 24 hours — notes, tripwires, re-issues, marker hygiene, latency, token spend, and the last error.

  • /jev:why

    Show the last few things the jev hooks said to Claude — the exact note or tripwire text, the signals behind it, and the marker text of any re-issue.

MCP servers (1)

  • jev

    runs: node ${CLAUDE_PLUGIN_ROOT}/dist/mcp.mjs

Hooks (9)

  • SessionStart

    node "${CLAUDE_PLUGIN_ROOT}/dist/hook.mjs" SessionStart

  • UserPromptSubmit
  • UserPromptSubmit

    node "${CLAUDE_PLUGIN_ROOT}/dist/hook.mjs" UserPromptSubmit --fallback

  • PreToolUse
  • PostToolUse
  • PostToolUse
  • PostToolUseFailure
  • Stop
  • SessionEnd
View source on GitHub

Similar plugins

CMEM Pro

claude

CMEM Pro for the Claude app - hooks stream tool use to cmem.ai and inject observations into new sessions and agents

3 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

mempalace

claudecodex

AI memory system — mine projects and conversations into a searchable palace. 47 MCP tools, auto-save hooks, guided setup.

3 skills · 6 commands · 1 MCP server · hooks

MemPalace/mempalace 60,536

pensyve

claude

Universal memory runtime for Claude Code with cross-session memory, entity-aware recall, lifecycle hooks, skills, commands, and memory-curator agents

7 skills · 7 commands · 2 agents · hooks

wshobson/agents 42,461

prowler

claude

Connect Claude to Prowler Cloud through the Prowler MCP server. Reads providers, scans, findings, compliance results and resources for AWS, Azure, GCP, Kubernetes, Microsoft 365, GitHub and the other providers listed at https://registry.prowler.com, and returns remediation guidance for failed checks. Writes are limited to Prowler Cloud itself: triggering scans, finding triage, mute rules and the mutelist, providers and integrations. Your cloud resources are not changed by the connector.

1 skill · 1 MCP server

prowler-cloud/prowler 15,234

worktrunk

claudecodex

Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows. This plugin provides configuration guidance (LLM commit messages, project hooks, worktree paths), automatic activity tracking (🤖/💬 indicators in `wt list` showing active Claude sessions), and a `/wt-diff` pane showing each worktree's changes.

2 skills · hooks

max-sixty/worktrunk 12,481

improve

claude

Point it at a codebase and it figures out what's worth doing — bugs, perf, tech debt, what to build next — then writes plans any agent can execute. It can hand the work to a cheaper model and review the result. It never edits your code.

1 skill

shadcn/improve 11,576

Maintain jev?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex pluginslisted