plumb-chat

A mailbox guide for talking with another agent working in the same plumb workspace. It explains how to address a live session, send notes, and interpret silence.

In plain words
What is it for?
Use it to find active sessions, warn another agent about files you plan to change, send answers, and understand mailbox limits and message expiry.
Why use it?
It reduces coordination conflicts when several agents may work on related files at the same time.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/plumbkit/plumb/plumb-chat
Any agent
npx skills add plumbkit/plumb --skill plumb-chat
Clone the repo
git clone --depth 1 https://github.com/plumbkit/plumb

Made for: Claude Code, Codex.

Per session 49 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,583 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00049 $0.01583
Opus 5 $0.00024 $0.00792
Sonnet 5 $0.00010 $0.00317
Haiku 4.5 $0.00005 $0.00158

Measured 3d ago against content hash 5a9b7fd82718, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

plumb-chat 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 3d 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.

internal/cli/skills/plumb-chat/SKILL.md · 73 lines

How it starts

The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Plumb's mailbox is two tools: leave_note sends, check_messages receives. It is for coordinating with a session working the same project right now — a claim on files you are about to touch, an answer only that agent has. Every message is agent-authored, secret-scrubbed before storage, and advisory: nothing sent here blocks anyone's write. Gated on [collab] mailbox, on by default and same-workspace only. Messages expire after [collab] intent_ttl_minutes, so an unread note does not sit forever.

Under a lean tool profile none of these tools is advertised (they stay callable by name); set [tools] profile = "full" in .plumb/config.toml to see them.

1. Address a live session, or the next one to attach

workspace_sessions()
leave_note(to="swift-falcon", body="rewriting the rate limiter — hold off on internal/tools/ratelimit")

to is a peer session's display NAME, which workspace_sessions lists alongside how long since each session's last tool call. A name is unique among LIVE sessions — plumb refuses to register or rename onto a name another live session holds, because the name is the address — but an ended session does not reserve it, so the same name can later belong to someone else.

Omit to and the message goes to the reserved "next": whoever attaches to this workspace next. That is same-project only and consumed by the first session to claim it — a handover note, not a broadcast.

When no live session answers to the name and no existing thread places it, the reply says so and files the message in THIS workspace's mailbox. It went nowhere else; do not read that as sent.

2. Reply into the thread you were given

leave_note(to="swift-falcon", conversation_id="c8f3a1b2c3d4e5f6", body="done — ratelimit is yours")

Every message carries a conversation_id. Sending without one mints a fresh thread and the reply tells you its id; quoting the id you were given puts your answer in the same thread.

3. Hand your turn over instead of polling

Read the full file on GitHub · 73 lines

Files

What ships with it

1 file 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.

  1. 3d ago First seen · 73 lines · 49 tokens per session scan A 5a9b7fd82718

Subscribe to this mod's changes

plumb-chat is a skill published in the GitHub repository plumbkit/plumb (4 stars, last pushed 3d ago), licensed MIT. It adds 49 tokens to every session and 1,583 once invoked, about $0.0002 per session on Opus 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 2026-08-31.

Related

Other skills, from other repositories