pebble-protocol

pebble-protocol is a skill for Claude Code, Codex from skyf0xx/Pebble. It costs 0 tokens per session (1,356 once invoked), scanned A, original, MIT.

A communication protocol for Pebble, a chat application with a minimal web interface. It requires the agent to send every reply through Pebble's messaging tool and supports plain messages or interactive controls.

In plain words
What is it for?
Use it when replying inside Pebble, especially for naming a session, sending conversational updates, or showing buttons and other interactive choices.
Why use it?
It prevents replies from being sent in a format Pebble cannot display. It also defines how the chat session gets a short label and how decisions or inputs should be presented.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it when replying inside Pebble, especially for naming a session, sending conversational updates, or showing buttons and other interactive choices.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/skyf0xx/pebble/pebble-protocol
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.

Any agent
npx skills add skyf0xx/Pebble --skill pebble-protocol
Clone the repo
git clone --depth 1 https://github.com/skyf0xx/Pebble

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 pebble-protocol

README.md
[![agentmods](https://agentmods.dev/badge/skills/skyf0xx/pebble/pebble-protocol.svg)](https://agentmods.dev/skills/skyf0xx/pebble/pebble-protocol)
Your own site
<a href="https://agentmods.dev/skills/skyf0xx/pebble/pebble-protocol"><img src="https://agentmods.dev/badge/skills/skyf0xx/pebble/pebble-protocol.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,356 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00000 $0.01356
Opus 5 $0.00000 $0.00678
Sonnet 5 $0.00000 $0.00271
Haiku 4.5 $0.00000 $0.00136

Measured 8d ago against content hash 770de73446ee, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

pebble-protocol 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 8d 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.

hermes-plugin/skills/pebble-protocol/SKILL.md · 154 lines

How it starts

The opening of the file, as written. The whole thing — 154 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Pebble Communication Protocol

You are connected to a user via Pebble — a warm, minimal PWA chat interface. All communication with the user MUST go through pebble_send. Never write a plain text reply directly.


The one tool: pebble_send

Every outbound message is a pebble_send call. The type field controls what Pebble renders.

type: "message" — plain text reply

Use for any conversational response, summary, or result that doesn't need interactive elements.

{
  "type": "message",
  "content": "Done. Here's what I found: ..."
}

Always set label on your first pebble_send of a session to name it in the session list. Pebble shows a provisional title taken from the user's first message until you do — a short (2–4 word) label that captures the task ("Fix deploy script", "Weekly report") replaces it and reads far better. Set it on the first reply; don't wait.

{
  "type": "message",
  "content": "On it.",
  "label": "Fix deploy script"
}

type: "ui" — interactive UI block

Use instead of asking the user to type a response. If you need a decision, confirmation, or input — render it. The user taps instead of types.

The spec is OpenUI Lang source (a string), not a JSON object. Each line is name = Expression; you must define root = Stack([...]); arguments are positional. A Button with no explicit action auto-sends its label back to you.

{
  "type": "ui",
  "spec": "root = Stack([prompt, actions])\nprompt = TextContent(\"Proceed with deploy?\", \"large-heavy\")\nactions = Buttons([yes, no])\nyes = Button(\"Yes, proceed\", null, \"primary\")\nno = Button(\"Cancel\", null, \"secondary\")"
}

When the user taps, the next turn arrives as a ui_action envelope:

{ "ui_action": "Yes, proceed", "payload": { "type": "continue_conversation", "params": {} }, "session_id": "<session_id>" }

ui_action is the tapped button's label (or a form's submit message); for a Form, the submitted field values arrive under payload.values keyed by field name. Act on it directly — do not re-ask.

Read the full file on GitHub · 154 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. 8d ago First seen · 154 lines · 0 tokens per session scan A 770de73446ee

Subscribe to this mod's changes

pebble-protocol is a skill published in the GitHub repository skyf0xx/Pebble (22 stars, last pushed 29d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,356 tokens. 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 2026-08-30.