manzanas: Skill for Claude Code

.agents/skills/broker-federation/SKILL.md

broker-federation is a skill for Claude Code, Codex from BariBariGood/manzanas. It costs 52 tokens per session (862 once invoked), scanned A, original, Apache-2.0.

A broker that presents several Manzanas simulator daemons through one address, including simulators reachable through an SSH tunnel. It handles finding a suitable simulator and assigning it temporarily to a client.

In plain words
What is it for?
Use it to list simulator hosts, find matching targets, create leases, and manage simulator work across a fleet of Macs from one broker endpoint.
Why use it?
It removes the need to search and schedule across multiple Macs yourself. After assignment, simulator actions, streams, state, recordings, and logs go directly to the assigned machine.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents).

This is BariBariGood/manzanas's own configuration. It tells Claude Code and Codex how to work on manzanas itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything manzanas configures →

Reuse

Borrowing it

Nothing to install: this file belongs to BariBariGood/manzanas. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/BariBariGood/manzanas/main/.agents/skills/broker-federation/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/BariBariGood/manzanas

Made for: Claude Code, Codex.

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.

agentmods badge for broker-federation

README.md
[![agentmods](https://agentmods.dev/badge/skills/baribarigood/manzanas/broker-federation/github.svg)](https://agentmods.dev/skills/baribarigood/manzanas/broker-federation)
Your own site
<a href="https://agentmods.dev/skills/baribarigood/manzanas/broker-federation"><img src="https://agentmods.dev/badge/skills/baribarigood/manzanas/broker-federation/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.

agentmods 80×15 button for broker-federation

Your own site · 80×15
<a href="https://agentmods.dev/skills/baribarigood/manzanas/broker-federation"><img src="https://agentmods.dev/badge/skills/baribarigood/manzanas/broker-federation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 52 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 862 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00052 $0.00862
Opus 5 $0.00026 $0.00431
Sonnet 5 $0.00010 $0.00172
Haiku 4.5 $0.00005 $0.00086

Measured 11d ago against content hash 4d2624a50be5, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

broker-federation scanned grade A with 1 finding 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 11d ago.

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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

the whole lease lifecycle; only raw curl/HTTP clients must re-point
.agents/skills/broker-federation/SKILL.md · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Broker federation

manzanas-broker fronts N daemons (one per Mac) for placement only: target enumeration + lease scheduling. After a grant, ALL target-bound work (boot, actions, streams, state, recording, journal) goes directly against the lease's host_addr — media and actions never flow through the broker. The manzanas CLI and MCP facade follow host_addr automatically, so you can keep them pointed at the broker address for the whole lease lifecycle; only raw curl/HTTP clients must re-point themselves.

Run it

Cross-platform; typically on a Linux box on the same tailnet:

make build
bin/manzanas-broker --addr :7440 \
  --host mac1=http://100.64.0.1:7433,intel \
  --host mac2=http://100.64.0.3:7433,arm64

curl -s localhost:7440/v0/healthz        # {"ok":true,"role":"broker","hosts":N}
curl -s localhost:7440/v0/fleet/hosts | jq   # per-host up/targets/active_leases

Alternatives: MANZANAS_BROKER_HOSTS='mac1=100.64.0.1:7433,intel;...' env (';'-separated) or --config fleet.json ({"hosts":[{"name","addr","labels"}]}). Names/addresses must be unique; static list — restart to add hosts. Run ONE broker per fleet.

Tunneled hosts

For a daemon that is not directly reachable from the broker's box (or bound to localhost), forward it over SSH and register the local end:

ssh -f -N -L 7439:localhost:7433 <user>@<mac-tailnet-ip>
bin/manzanas-broker --addr :7440 --host mac2tun=http://127.0.0.1:7439

Caveat: host_addr handed to clients is then http://127.0.0.1:7439, which only resolves on the broker's machine — either run clients there too, or give each client the same tunnel. Prefer direct tailnet addresses whenever possible.

Federated leasing

POST /v0/leases on the broker takes the normal AcquireLeaseRequest. Labels may include host-level labels (the host name and its configured extras): ["mac2","ios26"] pins the lease to one Mac; host labels are stripped before proxying.

  • 201 → active; the Lease carries host + host_addr — raw HTTP clients talk to host_addr from now on (the CLI/MCP do it for you).
  • 202 → queued on the least-loaded candidate; poll GET /v0/leases/{id} through the broker until active.
  • 409 no_match → no host has a matching target. 503 unavailable → all matching hosts are down (retryable outage, not a label mismatch).

Read the full file on GitHub · 76 lines

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.

  1. 11d ago First seen · 76 lines · 52 tokens per session scan A 4d2624a50be5

Subscribe to this mod's changes

broker-federation is a skill published in the GitHub repository BariBariGood/manzanas (20 stars, last pushed 10d ago), licensed Apache-2.0. It adds 52 tokens to every session and 862 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

ios-simulator-interaction

Interaction with the iOS simulator using iosef, a CLI optimized for agent usage. Use when building or testing changes on the iOS Simulator — viewing the screen, tapping buttons, reading accessibility trees, finding elements by selector, asserting UI state, scripting multi-step test flows, installing and launching…

riwsky/iosef · 100 tokens

testa

Autonomously E2E-test iOS apps in the Simulator — read the screen (accessibility tree OR on-device OCR), tap/type/swipe/drag-drop/pinch/rotate, manage apps, and assert results. Use when asked to test, QA, drive, or reproduce a flow in an iOS app/simulator (React Native, Expo, native SwiftUI, or any app).…

valewnrt/testa · 103 tokens

argent-test-ui-flow

Autonomously test an app UI (iOS or Android) by running interact-screenshot-verify loops using argent MCP tools. Use when testing UI flows, verifying login works, testing navigation, running end-to-end UI test scenarios, manual QA steps, visible UI changes, or visual behavior.

software-mansion/argent · 64 tokens

baguette

Drive iOS simulators programmatically via the baguette CLI — taps, swipes, multi-finger gestures, hardware buttons (Home / Lock / Volume / Action / Power), ASCII keyboard text, and frame capture, all without opening Xcode. Use when: (1) an agent needs to drive a booted iOS simulator from a script — tap a coordinate…

tddworks/baguette · 249 tokens

mobile-automation

Control Android and iOS devices, emulators and simulators — launch apps, tap, swipe, type, take screenshots, read the accessibility tree. Use when a task involves a mobile device or app, mobile UI testing, or reproducing a bug on a phone.

mobile-next/mobile-mcp · 58 tokens

argent-qa-flows

Create repeatable QA regression E2E tests as Argent flows from test cases, tickets, or acceptance criteria. Use when the user asks to generate or preserve an automated regression scenario, with deterministic setup, stable targets, executable structural or visual evidence, and two consecutive full passes. For one-off…

software-mansion/argent · 127 tokens