ai-maestro-agent-messaging
Send and receive messages between AI agents using AI Maestro's messaging system. Use this skill when the user asks to "send a message", "check inbox", "read messages", "notify [agent]", "tell [agent]", or any inter-agent communication.
What this skill does
# AI Maestro Agent Messaging
## Purpose
Enable communication between AI coding agents using AI Maestro's dual-channel messaging system. Agents are identified by their agent ID or alias, with tmux session names as a fallback. Supports both SENDING and RECEIVING messages.
## CRITICAL: Inter-Agent Communication
**YOU ARE AN AGENT** - This skill is for **agent-to-agent** communication, NOT human-agent communication.
### IMPORTANT: Understanding "Your Messages"
When the human operator says "check your messages" or "read your messages":
- **YOUR inbox** = Messages addressed TO YOUR AGENT (from anyone - operator, other agents, etc.)
- **NOT the operator's inbox** = You check YOUR inbox, not the operator's
**Example:**
- Human says: "Check your messages"
- You are agent: `backend-api`
- You check: `~/.aimaestro/messages/inbox/backend-api/` (YOUR inbox)
- These are messages addressed TO `backend-api` (from any sender)
- You DO NOT check: The operator's inbox or any other agent's inbox
### Agent Identity
- **Your inbox** = Messages addressed TO YOUR AGENT (from any sender)
- **Your agent ID** = Unique identifier for this agent (can also use agent name as fallback)
- **Your agent name** = The tmux session you're running in (get with `tmux display-message -p '#S'`)
- **Your inbox location** = `~/.aimaestro/messages/inbox/YOUR-AGENT-ID/` or `~/.aimaestro/messages/inbox/YOUR-AGENT-NAME/`
**You do NOT read:**
- ❌ The operator's inbox
- ❌ Other agents' inboxes
- ❌ Messages not addressed to your agent
**You DO read:**
- ✅ Messages addressed TO YOUR AGENT
- ✅ YOUR OWN inbox only
- ✅ Your agent's inbox: `~/.aimaestro/messages/inbox/YOUR-AGENT-ID/`
## When to Use This Skill
**Sending (Agent-to-Agent):**
- User (operator) says "send a message to [another-agent]"
- User says "notify [another-agent]" or "alert [another-agent]"
- User wants YOU to communicate with ANOTHER agent
- You need to send urgent alerts or requests to OTHER AGENTS
**Receiving (Check YOUR OWN Inbox):**
- User says "check my inbox" or "check my messages" = Use `check-aimaestro-messages.sh`
- User says "read my messages" or "read message X" = Use `read-aimaestro-message.sh <id>`
- User asks "any new messages?" = Use `check-aimaestro-messages.sh`
- Agent just started (best practice: check YOUR inbox first)
- You want to see what OTHER AGENTS have sent TO YOU
**RECOMMENDED WORKFLOW:**
1. First check for unread messages: `check-aimaestro-messages.sh`
2. Then read specific message: `read-aimaestro-message.sh <message-id>`
3. Message is automatically marked as read after reading
## Available Tools
## PART 1: RECEIVING MESSAGES (YOUR OWN INBOX)
**📖 QUICK START - Check and Read Messages:**
```bash
# Step 1: Check what unread messages you have
check-aimaestro-messages.sh
# Output shows:
# [msg-1234...] 🔴 From: backend-api | 2025-10-29 14:30
# Subject: Authentication endpoint ready
# Preview: The /api/auth/login endpoint is now...
# Step 2: Read the specific message (automatically marks as read)
read-aimaestro-message.sh msg-1234...
# Step 3: Check again - that message is now gone from unread
check-aimaestro-messages.sh
# Output: "📭 No unread messages"
```
**⚠️ CRITICAL: What "YOUR inbox" means:**
- YOU = The AI agent running in this tmux session
- YOUR inbox = `~/.aimaestro/messages/inbox/YOUR-AGENT-ID/` (or agent name as fallback)
- Messages in YOUR inbox = Messages OTHER AGENTS sent TO YOU
- NOT the operator's messages, NOT other agents' private messages
**IMPORTANT:** These commands check YOUR AGENT'S inbox only. They automatically:
1. Detect your current agent ID or agent name
2. Read from `~/.aimaestro/messages/inbox/YOUR-AGENT-ID/`
3. Show messages that OTHER AGENTS sent TO YOU
4. Do NOT access anyone else's inbox
### 1. Check YOUR Inbox for UNREAD Messages (Recommended)
**Command:**
```bash
check-aimaestro-messages.sh [--mark-read]
```
**What it does:**
- Shows ONLY UNREAD messages in YOUR inbox (messages sent TO YOUR AGENT)
- Automatically detects YOUR agent's session
- Displays: priority indicator, sender, subject, preview, timestamp
- Optional `--mark-read` flag to mark all messages as read after viewing
- **This is the recommended way to check messages** - avoids re-reading old messages
**Example:**
```bash
# Check unread messages without marking as read
check-aimaestro-messages.sh
# Check and mark all as read
check-aimaestro-messages.sh --mark-read
```
**Output format:**
```
📬 You have 3 unread message(s)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
[msg-167...] 🔴 From: backend-architect | 2025-10-29 13:45
Subject: API endpoint ready
Preview: The POST /api/auth/login endpoint is now...
[msg-168...] 🔵 From: frontend-dev | 2025-10-29 14:20
Subject: Need help with styling
Preview: Can you review the CSS for the navigation...
```
### 2. Read Specific Message and Mark as Read
**Command:**
```bash
read-aimaestro-message.sh <message-id> [--no-mark-read]
```
**What it does:**
- Retrieves and displays the full message content
- **Automatically marks the message as read** (unless `--no-mark-read` flag)
- Shows all message details: content, context, forwarding info
- Perfect for reading a specific message after checking the list
**Example:**
```bash
# Read message (automatically marks as read)
read-aimaestro-message.sh msg-1234567890-abc
# Peek at message without marking as read
read-aimaestro-message.sh msg-1234567890-abc --no-mark-read
```
**Output format:**
```
═══════════════════════════════════════════════════════════════
📧 Message: API endpoint ready
═══════════════════════════════════════════════════════════════
From: backend-architect
To: frontend-dev
Date: 2025-10-29 13:45:00
Priority: 🔴 urgent
Type: response
───────────────────────────────────────────────────────────────
The POST /api/auth/login endpoint is now deployed and ready...
───────────────────────────────────────────────────────────────
📎 Context:
{
"endpoint": "/api/auth/login"
}
✅ Message marked as read
═══════════════════════════════════════════════════════════════
```
### 3. Auto-Display on Agent Start (Legacy - DO NOT USE MANUALLY)
**Command:**
```bash
check-and-show-messages.sh
```
**What it does:**
- Automatically runs when you attach to a tmux session
- Shows a summary of unread messages
- **DO NOT run this command manually** - it's for auto-display only
- **For manual checking, use `check-aimaestro-messages.sh` instead**
**Why not use this manually?**
- It's designed for auto-display (runs on tmux attach)
- Output format is optimized for quick glance, not interactive reading
- Use the new commands (#1 and #2 above) for better experience
**Output format:**
```
Message: msg_1234567890_abcde
From: backend-architect ← Another agent sent this TO YOU
To: frontend-dev ← YOUR session name
Subject: Need API endpoint
Priority: high
Type: request
Status: unread
Timestamp: 2025-01-17 14:23:45
Content: Please implement POST /api/users with pagination...
```
### 4. Check for New Messages Count (Quick)
**Command:**
```bash
check-new-messages-arrived.sh
```
**What it does:**
- Shows count of unread messages in YOUR inbox
- Automatically checks YOUR session's inbox
- Quick check without full details
- Returns "No new messages" or "You have X new message(s)"
**Example:**
```bash
check-new-messages-arrived.sh
# Output: "You have 3 new message(s)" ← Messages sent TO YOU
```
### 5. Read Specific Message FROM YOUR Inbox (Direct File Access - Advanced)
**Command:**
```bash
cat ~/.aimaestro/messages/inbox/$(tmux display-message -p '#S')/<message-id>.json | jq
```
**What it does:**
- Read a specific message file from YOUR inbox
- `$(tmux display-message -p '#S')` = YOUR session name (auto-detected)
- Use `jq` for pretty formatting
- Useful when you know the message ID
**Directory structure:**
```
~/.aimaestro/messages/
├── inbox/YOUR-SESSION-NAME/ # Messages TO YOU from other agents
│ └── msg_*.json
├── sent/YOUR-SESSION-NAME/Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.