An orchestrator for guided coding sessions where a non-engineer describes changes in chat or by marking up a running web app. Coding workers make the changes in an isolated worktree, a separate copy of the repository for that work.

It helps start or resume isolated local web and desktop app sessions, dispatch coding workers, inspect the running app, and coordinate changes for later handoff to engineering.

It lets the person describing the change avoid using Git or editing source code directly while keeping the resulting work separated from the main codebase. The orchestrator coordinates the session rather than making code changes itself.

vibe — Skill for Claude Code and Codex by closedloop-ai/claude-plugins

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add closedloop-ai/claude-plugins --skill vibe

Clone the repo

git clone --depth 1 https://github.com/closedloop-ai/claude-plugins

Made for: Claude Code, Codex.

Or install vibe, the plugin that ships this one along with the rest of its 10 skills, 16 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/closedloop-ai/claude-plugins/vibe/github.svg)](https://agentmods.dev/skills/closedloop-ai/claude-plugins/vibe)

Your own site

<a href="https://agentmods.dev/skills/closedloop-ai/claude-plugins/vibe"><img src="https://agentmods.dev/badge/skills/closedloop-ai/claude-plugins/vibe/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/closedloop-ai/claude-plugins/vibe"><img src="https://agentmods.dev/badge/skills/closedloop-ai/claude-plugins/vibe.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 152 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 8,788 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00152	$0.08788
Opus 5.5	$0.00061	$0.03515
Sonnet 5.5	$0.00030	$0.01758
Haiku 4.5	$0.00015	$0.00879

Measured today against content hash 2fcb213170c5, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

vibe scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.

The scan reads SKILL.md. This mod also ships 17 executable files (scripts/commit-worktree.mjs, scripts/commit-worktree.test.mjs, scripts/dist/claude-worker.mjs, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/vibe/skills/vibe/SKILL.md · 605 lines

How it starts

The opening of the file, as written. The whole thing — 605 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Vibe

You are the orchestrator of a vibe session. You pair with a person who is not an engineer and does not use git. They describe what they want in chat or by annotating the app on their Vercel environment; workers you dispatch make the change in code, and the person sees it when they ask you to redeploy. Everything produced is handed to design and then engineering, so it must already look like code the team would write.

Your role: orchestrate, never do the work

A session can run for hours. Your context is for the conversation with the person and for keeping track of the session, so you NEVER spend it on the work itself. Under all circumstances:

You never read source files, search the codebase, explore the repo, edit or create files, read tickets in full, or run builds, tests, linters, installers, git, or gh.
You commit; no worker ever does, because a commit runs the repository's commit hooks. You commit only through scripts/commit-worktree.mjs, which returns short JSON.
You may: talk to the person; run this skill's own scripts (scripts/vibe-preflight.sh, scripts/vibe-sessions.mjs, scripts/commit-worktree.mjs, scripts/local-plans.mjs, scripts/dist/writer-state.mjs, scripts/dist/claude-worker.mjs), whose output is short JSON; call ClosedLoop get-me and closedloop-graph sync_status to check the connectors; open, reload, and inspect the in-app browser; and dispatch workers.
Everything else goes to a worker, even a one-line change and even when you think you already know the file. If you notice yourself about to open a source file, dispatch a worker instead.
Give each worker only what it needs: the worktree path, the session summary and mode, the live ticket slug, the request in the person's own words, and for an annotation the comment text, the element context, and the page route. Workers return a short result; do not ask them for file contents.

Follow references/quality-loop.md for the one persistent writer, serial queue, internal planning/review and communication. The person never sees a technical plan or technical question, and needs no upfront summary. Updates say only which feature is complete and what is being worked on next. Research product questions through closedloop-graph and live evidence first; interrupt only for an absolutely necessary unresolved product question, one at a time with full context. Never re-ask a settled decision.

Read the full file on GitHub · 605 lines

Files

What ships with it

29 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

today Changed · +46 lines 2fcb213170c5
yesterday Changed · +111 lines · -180 tokens per session 30881618a28a
2d ago Changed · +211 lines · +135 tokens per session 0a7eca2a6f35
3d ago Changed · +24 lines 4ebea50ab372
4d ago First seen · 213 lines · 197 tokens per session scan A d2d77354dd49

Subscribe to this mod's changes

vibe is a skill published in the GitHub repository closedloop-ai/claude-plugins (122 stars, last pushed today), licensed Apache-2.0. It adds 152 tokens to every session and 8,788 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-06.

Claude Code Frontend LLM Project management agentic-ai ai-agents anthropic developer-tools

file-a-task

File work into nohuman (taskadd) and check on it (taskstatus) via the nohuman MCP bridge, instead of doing the work inline.

no-human-ai/no_human · 37 tokens

fleet

The session roster (/foundry:fleet) — native /list-agents annotated with each row's foundry programme, over the NATIVE session/container list. Read-only/advisory. Trigger to see all your parallel sessions and containers at a glance with their foundry context, or when supervising many sessions and asking "what is each…

lukasrepublic/agentic-foundry · 73 tokens

parallel-feature-development

Coordinate parallel feature development with file ownership strategies, conflict avoidance rules, and integration patterns for multi-agent implementation. Use this skill when decomposing a large feature into independent work streams, when two or more agents need to implement different layers of the same system…

wshobson/agents · 105 tokens

bazel-build-optimization

Optimize Bazel builds for large-scale monorepos. Use when configuring Bazel, implementing remote execution, or optimizing build performance for enterprise codebases.

wshobson/agents · 36 tokens

company-ceo

Run a PenguinHarness organization as its CEO — turn the mission into a ticket tree, hire HR and finance first, partition the shared workspace, schedule the calendar, open a channel per stream, review tickets and report to the board in the all-hands channel.

Prism-Shadow/penguin-harness · 56 tokens

sdlc-accelerate

End-to-end SDLC ramp-up from idea to construction-ready with automated phase transitions.

jmagly/aiwg · 23 tokens