claude-mem
claudecodexPersistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
Jev judgment tools and the Jev agent skill.
Not in a marketplace yet, so it loads for one session with --plugin-dir.
Plugin jev in panaalexandrucristian/jev-mcp-private. Commands read from the repo's manifest — 26 days ago.
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.
releasingReleases this package to npm and GitHub. Use when cutting a new version of @jkudish/jev-mcp, bumping versions, writing changelogs, publishing to npm, tagging releases, or troubleshooting a publish that did not go as expected.
releasing-private-pluginReleases a new version of the private jev plugin from this repository (version bump in 4 files, PRIVATE.md paragraphs, release commit, tag, push and CI check). Use when the owner asks to release, ship, publish, tag or bump the version of the plugin, or says facem release. Not for the npm package of
handoff-verifyVerify a handoff/handover note against the session transcript for lost details and wrong facts. Use automatically whenever the user asks to write, prepare, update or check a handoff, handover, CONTINUE-HERE note, or notes for the next session (e.g. "write a handoff", "create a HANDOVER.md", "wrap up
jev-control-modePuts an explicitly enabled Claude Code session under Jev control: every choice with 2+ real alternatives (approach, task order, which files to search or read, commands and tests, edit variants, delegation and model, when to ask the user, when the work is done) is decided by Jev judgments at a config
jev-flowOpt-in adaptive coding workflow (active only with JEV_FLOW=on, or when the user asks for jev-flow) for non-trivial code tasks in a repository (debugging, implementation, refactoring, finishing a change): delegate broad code discovery to the jev-locator subagent instead of searching in the main threa
jevConventions for the Jev judgment tools. Use when screening fetched or pasted content, verifying claims against evidence, ranking or classifying items by meaning, comparing passages, extracting fields, reviewing a patch, gating completion, or judging how likely a proposition is — and when choosing be
/jev:jev-controlPut this session under Jev control (on/off/status/threshold/gate). Every choice with two or more real alternatives is then decided by Jev at a configurable confidence (default 0.95).
/jev:jev-doneFinish a code change with jev-flow - the gate runner runs the real checks, pairs every claim with its real evidence and calls jev_gate itself on every part of the final diff; you write claims, read one aggregated report and answer once, never payloads.
/jev:jev-locateDelegate a code-location question to the read-only jev-locator subagent and get compact, hash-checked locations back.
/jev:prompt-checkSwitch the prompt check on or off for this session (on/off/status/threshold/lang). When on, Jev checks each later prompt and shows the user one short tip line only if it is at least `threshold` sure of a problem.
handoff-readerRead-only reader for the delegated review of handoff-verify. Use ONLY when the user explicitly asked for delegated review and the parent built an assignment with delegation.py; it reads the assigned prepared chunk files and returns candidate details with exact quotes as one JSON object. It never cal
jev-locatorRead-only code locator for jev-flow. Give it one question about where behavior lives in the current repository; it runs the local candidate helper, which applies the ranking rule itself (jev_rerank or jev_find with top_k 5 when candidates are ambiguous) and prints only compact ranked hits, verifies
jevruns: npx -y --package=@jkudish/jev-mcp@latest jev-mcp
node "${CLAUDE_PLUGIN_ROOT}/scripts/jev-flow-hook.mjs" SessionStart
node "${CLAUDE_PLUGIN_ROOT}/scripts/jev-flow-hook.mjs" UserPromptSubmit
node "${CLAUDE_PLUGIN_ROOT}/scripts/jev-flow-hook.mjs" PreToolUse
node "${CLAUDE_PLUGIN_ROOT}/scripts/jev-flow-hook.mjs" PostToolUse
node "${CLAUDE_PLUGIN_ROOT}/scripts/jev-flow-hook.mjs" PostToolUseFailure
node "${CLAUDE_PLUGIN_ROOT}/scripts/jev-flow-hook.mjs" Stop
node "${CLAUDE_PLUGIN_ROOT}/scripts/jev-flow-hook.mjs" SubagentStop
node "${CLAUDE_PLUGIN_ROOT}/scripts/jev-flow-hook.mjs" SessionEnd
Problems found in this repo's manifests
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
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
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
8 skills · 1 MCP server
Code intelligence powered by a knowledge graph. Provides execution flow tracing, blast radius analysis, and augmented search across your codebase.
12 skills · 1 MCP server · hooks
daisyUI Blueprint MCP server - (Needs paid license)
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.