Lody is a shared workspace that connects coding agents running on different computers and lets people view and direct their work from desktop, mobile, web, or a command line. Teams use it to share agent conversations, monitor progress and code changes, and assign work to connected machines while retaining their existing agents and configurations.
Borrowing it
Nothing to install: this file belongs to LodyAI/Lody. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/LodyAI/Lody/main/.claude/skills/lody-loro-sync-stack/SKILL.mdgit clone --depth 1 https://github.com/LodyAI/LodyWrote 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/skills/lodyai/lody/lody-loro-sync-stack)<a href="https://agentmods.dev/skills/lodyai/lody/lody-loro-sync-stack"><img src="https://agentmods.dev/badge/skills/lodyai/lody/lody-loro-sync-stack/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/skills/lodyai/lody/lody-loro-sync-stack"><img src="https://agentmods.dev/badge/skills/lodyai/lody/lody-loro-sync-stack.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 | $0.00089 | $0.02283 |
| Opus 5.5 | $0.00036 | $0.00913 |
| Sonnet 5.5 | $0.00018 | $0.00457 |
| Haiku 4.5 | $0.00009 | $0.00228 |
Grade A, and why
lody-loro-sync-stack 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 20d 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.
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 — 155 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Lody Loro sync stack
Two planes, and they are not interchangeable
Lody moves state over Loro Streams on planes with different guarantees. Mixing them is the source of the worst bugs in this subsystem.
Durable documents — session history, workspace meta, tasks, preview comments.
Ordered, persisted, replayed from an offset, and never dropped. Owned by
StreamsTransportAdapter (apps/cli/src/lib/loro/streams-transport.ts). If a
reader must receive something, it goes here.
Ephemeral channels — current liveness only. Not persisted with the document stream, no history replay, latest-state oriented. A server may keep only a small live fan-out queue and disconnect a slow subscriber rather than buffer every intermediate update. Everything here is disposable by design.
Two ephemeral channels exist, on separate stream URLs and therefore separate transports and separate queues:
| Channel | Constant | Carries |
|---|---|---|
presence |
LODY_PRESENCE_CHANNEL (packages/shared/src/presence.ts) |
machine heartbeat, session active status, session-viewing |
machine-monitor |
packages/shared/src/machine-monitor.ts |
observer-leased device resource snapshots |
Keep them separate. Machine-monitor snapshots are far larger and more frequent than anything the presence channel may carry; merging the two would put that traffic in front of the machine heartbeat.
Healthy durable sync proves nothing about presence delivery. They are different transports. A machine can sync conversations perfectly and still read offline.
The invariant
Publish only small, low-frequency liveness state on an ephemeral channel. Never publish high-frequency or high-volume data on it.
This is binding. The full statement is
specs/loro-ephemeral-presence-channel.md;
the scoped rules are in
apps/cli/src/lib/loro/AGENTS.md.
Why it bites
One workspace owns ONE presence transport, and the machine heartbeat shares it with
every other presence writer in that workspace. Publication is strictly serial: one
update in flight. In consumed streams-crdt 0.16.2, the built-in adaptor coalesces
unsent single-key set/delete updates; it never replaces in-flight bytes or merges
distinct/unknown/multi-key updates. There is no heartbeat priority or distinct-key
traffic bound. pendingLocalCount counts logical writes, including superseded
ones, rather than POSTs. A writer that
produces faster than the network drains delays every later writer.
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.
- 20d ago Changed · +4 lines 57a6a0514c83
- 1mo ago First seen · 151 lines · 89 tokens per session scan A 43b0fd802e53
lody-loro-sync-stack is a skill published in the GitHub repository LodyAI/Lody (1,640 stars, last pushed today), licensed Apache-2.0. It adds 89 tokens to every session and 2,283 once invoked, about $0.0004 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 2026-09-21.
Other skills, from other repositories
room-playbook
Lightweight default rules for coordinating a Room conversation without confusing chat wakeups with durable WorkGraph responsibility.
webapp-building
Build or modify production-oriented web applications by following the repository architecture and validating real user flows.
access-exchange-2010-ews-functions
Exchange 2010 EWS integration for emails, calendar, contacts, and tasks.
access-zyla-api-hub
Zyla API Hub Skill — Turn your OpenClaw AI agent into a real-world operator. Power it with 10,000+ production-ready APIs from Zyla API Hub — instant access to weather, finance, translation, email validation, geolocation, and more.
alibaba-cloud-chatbot-manager
Manage Alibaba Cloud beebot (Chatbot) via OpenAPI/SDK. Use whenever the user asks to configure, query, or troubleshoot Alibaba Cloud chatbot resources, including bot inventory, configuration changes, status checks, and API-level diagnostics.
vigilante-issue-implementation-on-php
Implement a GitHub issue end-to-end when Vigilante dispatches work for a PHP repository with Composer, static analysis, and security guidance.