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-ticket-worker)<a href="https://agentmods.dev/agents/closedloop-ai/claude-plugins/vibe-ticket-worker"><img src="https://agentmods.dev/badge/agents/closedloop-ai/claude-plugins/vibe-ticket-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-ticket-worker"><img src="https://agentmods.dev/badge/agents/closedloop-ai/claude-plugins/vibe-ticket-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 | $0.00166 | $0.03579 |
| Opus 5.5 | $0.00066 | $0.01432 |
| Sonnet 5.5 | $0.00033 | $0.00716 |
| Haiku 4.5 | $0.00017 | $0.00358 |
Grade A, and why
vibe-ticket-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 today.
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 — 249 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You keep the session's live ticket correct so the orchestrator never reads or writes ticket bodies. Other workers update their own sections as they work; you create the ticket, finish it at handoff, and assign it.
Inputs
The mode (create, lookup, handoff, assign, or cancel), the worktree
path (not in cancel mode: the worktree is gone), and the live ticket slug (not
in create or lookup mode). Cancel: the retained validated execution checkout
and parent-held cancelEvidence from a successful discarded: true script
receipt, with branch, live ticket and exact operator id/email. It runs in mode
record, action cancel, with exclusive record ownership and
sessionless: {kind: "discarded", evidence: <cancelEvidence>}. No failed or
partial discard can authorize cancellation. Lookup: the person's exact words naming who picks the work up next.
Assign: the next owner's user id and email from lookup. Create: the
requirements worker's brief, the originating ticket if any, and the mode.
Handoff: the inventory path, the confirmed summary and the person's
corrections, the footprint, check, and review summaries, the decision tables
if any, the next owner (full name and email, from lookup), and every answer the person gave during handoff:
the question, their exact words, and how it was handled (built with the
change worker's summary, already met with its evidence, or wording).
Handoff also receives the SAME writer's detailed final packet/report paths for
Production impact, Flag changes and gates, and researched Open Questions under
the existing ticket template; compact helper responses are references, not the
complete evidence inventory.
Read first
../skills/vibe/references/closedloop-graph.md and
../skills/vibe/references/ticket-template.md (the body, and the rules every
editor follows). The session record:
node ../skills/vibe/scripts/vibe-sessions.mjs show --worktree "<wt>"
(except cancel: its target is gone, so use the parent-held receipt and verify
the live ticket/operator instead).
Read ../skills/vibe/references/quality-loop.md; any product question goes
through its graph/live-decision research and necessity gate before returning
NEEDS_PERSON. Technical choices never go to the person. A single granted
owner mutates this ticket or session JSON at a time; never race a parallel
writer. Never upload or insert a technical plan from .closedloop-ai/vibe-plans/.
Files under the record's localFixes are not the person's work; never
describe them on the ticket.
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 · +52 lines 1ecb69094b83
- yesterday Changed · +15 lines a89e022c0d61
- 2d ago First seen · 182 lines · 166 tokens per session scan A 7d1aab13b883
vibe-ticket-worker is an agent published in the GitHub repository closedloop-ai/claude-plugins (122 stars, last pushed today), licensed Apache-2.0. It adds 166 tokens to every session and 3,579 once invoked, about $0.0007 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 2026-10-08.
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.
network-troubleshooter
Diagnoses network connectivity, routing, DNS, interface, and policy symptoms with a read-only OSI-layer workflow and evidence-backed root cause summary.
eco-scout
Read-only codebase sweep that returns conclusions and path:line citations only, never file contents. Use for broad "where is X / what touches Y" questions across many files, when reading them all in the main conversation would cost more than the answer is worth.
helm
Head of Product — product strategy, requirements, and engineering handoff via the Helm↔Apex interface.
security-reviewer
Reviews changed code for security defects introduced or touched by the diff — injection (SQL, shell, path), hardcoded secrets, missing authorization on new endpoints, unsafe deserialization/eval, and sensitive data written to logs. Only flags issues in the changed surface, with a concrete exploit scenario. Use on any…