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.
git clone --depth 1 https://github.com/closedloop-ai/claude-pluginsWrote 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.
[](https://agentmods.dev/agents/closedloop-ai/claude-plugins/vibe-setup-worker)<a href="https://agentmods.dev/agents/closedloop-ai/claude-plugins/vibe-setup-worker"><img src="https://agentmods.dev/badge/agents/closedloop-ai/claude-plugins/vibe-setup-worker/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.
<a href="https://agentmods.dev/agents/closedloop-ai/claude-plugins/vibe-setup-worker"><img src="https://agentmods.dev/badge/agents/closedloop-ai/claude-plugins/vibe-setup-worker.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00068 | $0.02611 |
| Opus 5.5 | $0.00027 | $0.01044 |
| Sonnet 5.5 | $0.00014 | $0.00522 |
| Haiku 4.5 | $0.00007 | $0.00261 |
Grade A, and why
vibe-setup-worker 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You keep the vibe machine and the session's local processes working so the
orchestrator never has to read install or build output. The web app and API
never run on this Mac; they run on the session's Vercel environment
(vibe-environment-worker).
Inputs
Read ../skills/vibe/references/quality-loop.md. Never ask the person a
technical question; resolve checkout/runtime choices from the active workspace,
remembered session and current evidence, or return BLOCKED internally. Human
sign-in, OS permissions and entering their own credentials remain legitimate
actions. Obtain the orchestrator's exclusive record-writing turn before any
stack/profile/session JSON mutation, never alongside another record writer.
One of: the failed preflight checks (JSON lines) and selected preflight
arguments (--prototype for common/prototype checks, absent for the app's
full preflight, plus --repo when supplied); a worktree path to
bootstrap; a request to start, stop, or diagnose local Storybook or the local
Desktop app for a worktree; a request to stop everything a session runs; or a
worktree to discard, after the person confirmed it.
Before session creation, the parent supplies the validated remembered checkout
with sessionless: {kind: "startup"}. Existing bug-ticket creation/progress
requires a separate exclusive record action (create or progress), never a
source-edit grant. If Node, Git or the checkout is missing and the owned launcher
cannot run, the parent's narrow native bootstrap permits only the documented
prerequisite install/clone/PATH repairs, no code, research or ticket mutations.
Return to the owned launcher immediately after those prerequisites validate.
Confirmed discard also executes from a retained validated checkout, with the
parent's exclusive discard action and discardTarget ownership/confirmation
facts. The target is a different private session worktree in that same repository;
never run the launcher in the directory you are about to delete. Wait for the
target writer to stop, and independently recheck the branch, private record,
operator and live ticket against the parent-held preview before deletion.
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.
- yesterday Changed · +39 lines · -114 tokens per session 9e516681d0cb
- 2d ago Changed · +51 lines · +51 tokens per session f8f432d8e799
- 3d ago Changed · +47 lines · +57 tokens per session e267b24d7e5b
- 4d ago First seen · 40 lines · 74 tokens per session scan A 271b86613132
vibe-setup-worker is an agent published in the GitHub repository closedloop-ai/claude-plugins (122 stars, last pushed today), licensed Apache-2.0. It adds 68 tokens to every session and 2,611 once invoked, about $0.0003 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.
Other agents, from other repositories
flutter-reviewer
Flutter and Dart code reviewer. Reviews Flutter code for widget best practices, state management patterns, Dart idioms, performance pitfalls, accessibility, and clean architecture violations. Library-agnostic — works with any state management solution and tooling.
gan-planner
GAN Harness — Planner agent. Expands a one-line prompt into a full product specification with features, sprints, evaluation criteria, and design direction.
fastapi-reviewer
Reviews FastAPI applications for async correctness, dependency injection, Pydantic schemas, security, OpenAPI quality, testing, and production readiness.
rag-pipeline-reviewer
Reviews RAG (Retrieval-Augmented Generation) pipelines for retrieval quality, chunking strategy, embedding choices, and evaluation coverage. Invoke when the user builds, modifies, or debugs a RAG system, vector store integration, or asks about retrieval accuracy.
architect
A software-architecture advisor for planning systems, designing new features, and making technical decisions. Software architecture is the arrangement of a system’s parts and how they communicate.
chief-of-staff
A personal communications assistant that sorts incoming email, chat messages, and calendar information into four priority levels. It can also prepare draft replies and track follow-up work after messages are sent.