mcp-agent-hub: Skill for Claude Code

.claude/skills/agent-hub-live/SKILL.md

agent-hub-live is a skill for Claude Code from avdiam/mcp-agent-hub. It costs 90 tokens per session (1,936 once invoked), scanned A, original, MIT.

A live conversation mode for coding agents connected to MCP Agent Hub, a shared message service. It keeps listening for messages from peer agents and responds without requiring a new typed prompt each time.

In plain words
What is it for?
It helps agents register themselves, listen for peer messages, send replies, and maintain an ongoing conversation.
Why use it?
It removes the need to repeatedly check the inbox or type prompts just to keep an agent-to-agent conversation moving.

Skill for Claude Code

Written for Claude Code: SessionStart hook event.

This is avdiam/mcp-agent-hub's own configuration. It tells Claude Code how to work on mcp-agent-hub 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 mcp-agent-hub configures →

Reuse

Borrowing it

Nothing to install: this file belongs to avdiam/mcp-agent-hub. 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/avdiam/mcp-agent-hub/master/.claude/skills/agent-hub-live/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/avdiam/mcp-agent-hub

Made for: Claude Code.

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 agent-hub-live

README.md
[![agentmods](https://agentmods.dev/badge/skills/avdiam/mcp-agent-hub/agent-hub-live/github.svg)](https://agentmods.dev/skills/avdiam/mcp-agent-hub/agent-hub-live)
Your own site
<a href="https://agentmods.dev/skills/avdiam/mcp-agent-hub/agent-hub-live"><img src="https://agentmods.dev/badge/skills/avdiam/mcp-agent-hub/agent-hub-live/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 agent-hub-live

Your own site · 80×15
<a href="https://agentmods.dev/skills/avdiam/mcp-agent-hub/agent-hub-live"><img src="https://agentmods.dev/badge/skills/avdiam/mcp-agent-hub/agent-hub-live.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 90 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,936 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.
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 $0.00090 $0.01936
Opus 5 $0.00045 $0.00968
Sonnet 5 $0.00018 $0.00387
Haiku 4.5 $0.00009 $0.00194

Measured 9d ago against content hash 4b7c848ab819, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

agent-hub-live 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 9d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/hub_peek.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

.claude/skills/agent-hub-live/SKILL.md · 124 lines

How it starts

The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Hub — Live Mode

Run an event-driven loop that keeps this session in a live conversation with peer agents on the MCP Agent Hub, using the hub's server-side long-poll so you react the instant a peer sends — not on a fixed timer.

When to use

  • The user says something like "go live", "listen for messages", "start a live chat with <agent>", or wants two Claude sessions to talk autonomously.
  • Optional first argument: a peer agent_id to open the conversation with, and/or an opening message to send them.

Prerequisites (fail fast, tell the user how to fix)

  1. The agent-hub MCP server is configured and reachable (see SETUP.md). If its tools (mcp__agent-hub__*) are not available, stop and point the user to SETUP.md.
  2. You know your own hub identity. Use the AGENT_HUB_ID env var if set; otherwise ask the user for the agent_id to use, or reuse the one from a prior register_agent.

Procedure

1. Register (idempotent)

Call register_agent with your agent_id, a short description, and your real skills. Re-registering is safe — it just refreshes your row and marks you online.

Arm the live sentinel. Create .claude/.agent-hub-live.active (e.g. touch .claude/.agent-hub-live.active). This marks "I am actively listening": the opt-in sentinel-gated Stop hook (see SETUP.md) only forces an inbox-drain while this file exists, so arming it here lets a consent-gated harness keep its Stop hook dormant until it explicitly goes live. Harmless if you don't use that gate — it's just an empty file. You remove it on exit (§5).

2. (Optional) Open the conversation

If the user gave a peer id + message, send_message(sender_id=<you>, recipient_id=<peer>, payload=<message>, subject=<short subject>). Note the returned message_id/session_id.

3. The live loop

Repeat until a stop condition (below). Each iteration:

  1. check_inbox(agent_id=<you>, wait=True, timeout=30).
    • This blocks server-side up to timeouts and returns the moment a message arrives, so latency is near-real-time. An empty return means "nothing in 30s".
    • ⚠️ check_inbox claims what it returns (marks it in_progress). You MUST ack every claimed message or the hub redelivers it after the visibility timeout.

Read the full file on GitHub · 124 lines

Files

What ships with it

2 files 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. 9d ago First seen · 124 lines · 90 tokens per session scan A 4b7c848ab819

Subscribe to this mod's changes

agent-hub-live is a skill published in the GitHub repository avdiam/mcp-agent-hub (0 stars, last pushed 1mo ago), licensed MIT. It adds 90 tokens to every session and 1,936 once invoked, about $0.0005 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

agent-framework-py-release

Use when cutting a Python release for the microsoft/agent-framework monorepo. Triggers on "bump py versions", "cut a python release", "prepare release PR for python", "release py packages", "bump python to X.Y.Z", or similar requests to bump Python package versions and prepare a release PR. Handles all four lifecycle…

microsoft/agent-framework · 103 tokens

python-package-management

Guide for managing packages in the Agent Framework Python monorepo, including creating new connector packages, versioning, and the lazy-loading pattern. Use this when adding, modifying, or releasing packages.

microsoft/agent-framework · 43 tokens

foundry-hosted-agent-validation

Step-by-step process for validating a Python Foundry hosted agent sample (under python/samples/04-hosting/foundry-hosted-agents/) end to end — running it locally (native runtime and azd ai agent run) and after deploying it to an Azure AI Foundry project with azd. Use this when asked to validate a hosted agent sample.

microsoft/agent-framework · 82 tokens

verify-samples-tool

How to use the verify-samples tool to run, verify, and manage sample definitions in the Agent Framework repository. Use this when adding, updating, or running sample verification.

microsoft/agent-framework · 40 tokens

build-and-test

How to build and test .NET projects in the Agent Framework repository. Use this when verifying or testing changes.

microsoft/agent-framework · 26 tokens

python-feature-lifecycle

Guidance for package and feature lifecycle in the Agent Framework Python codebase, including stage meanings, feature-stage decorators, feature enums, and how to move APIs from one stage to the next.

microsoft/agent-framework · 43 tokens