One-click install
npx skills add https://github.com/grandamenium/cortextos --skill comms-grandamenium
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: comms
Source: https://github.com/grandamenium/cortextos/tree/main/templates/analyst/.claude/skills/comms
Command: npx skills add https://github.com/grandamenium/cortextos --skill comms-grandamenium

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents delayed or mishandled communication inside an agent session by ensuring every injected Telegram or agent message is read and responded to immediately using the correct bus command and threading identifiers.

Core Features & Use Cases

  • Real-time message handling: Processes injected blocks from the fast-checker daemon, including Telegram messages and inter-agent communications.
  • Correct reply routing: Uses the required send-telegram and send-message commands, including msg_id for agent replies to maintain conversation continuity.
  • Priority-aware workflow: Guides the agent to handle urgent items first, process callback data, and correctly deal with photo inputs by using provided local paths.
  • Turn control to avoid queueing: Ensures that when a user asks a question requiring immediate response, the agent stops tool execution and ends its turn so the reply is delivered next.

Quick Start

Use the comms Skill to reply to an incoming Telegram block by sending the response exactly with the command shown in that message header.

Frequently Asked Questions about comms

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I handle incoming Telegram messages inside an agent session without delaying workflow execution?

Telegram message handling inside an agent session uses routing commands to parse injected message blocks and respond immediately. The agent applies priority-aware workflow logic, processes callbacks, and sends replies using the correct bus command to maintain conversation continuity.

What is the best way to route inter-agent communication and approval workflows in real-time?

Real-time inter-agent communication routing parses message headers and applies the correct bus send command for each message type. Agent replies are threaded via msg_id to ensure conversation continuity across approval workflows and callback coordination.

How do I thread agent replies to maintain conversation continuity when responding to injected messages?

To maintain conversation continuity, agent replies are threaded using the msg_id identifier from the original injected message. The send-message command links the response to the active conversation block within the cortextOS session.

When should I use turn control and end the current agent turn during a live session?

Turn control is required when a user asks a question needing an immediate response during a live session. The agent stops tool execution and ends its turn so the reply is delivered next, preventing message queueing behind active workflows.

Can I process photo inputs and callback data within Telegram message routing workflows?

Yes, Telegram message routing workflows process callback data and handle photo inputs by utilizing the local file paths provided within the injected message blocks, ensuring the agent accesses media correctly before resuming other work.

Why does my agent queue user responses instead of replying immediately during tool execution?

Agent responses queue because the current turn is not ended when awaiting user input. Stopping tool execution and ending the turn immediately after parsing the message ensures the reply is delivered next instead of being delayed behind other operations.