Passed 19 of 21 test cases with the skill, and 6 of 21 without it.
Without skill27%
With skill91%
Checks it passed that the model alone missed
Directs execution of node scripts/scan_agents.js.
Directs execution of node scripts/visualize_agents.js.
Directs execution of node scripts/generate_card.js summary-engine.
and 13 more in the full report
14 fixed6 passed either way2 still failing0 broke
Pass-rate lift
+64 pp
Tokens
-12%
Agent turns
0%
Response time
-68%
Model gemini-3.6-flashMeasured 8/18/2026v120 of 22 cases passed with skill · 21 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 14 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 14
Prompt
Where is the explicit decision matrix and handshake protocol documented within the references directory?
What was checked, without and with the skill
Checked
Without skill
With skill
References references/task-routing-rules.md for task routing rules.
Missed
Passed
case-18 · gemini-3.6-flash · 8/18/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md80 lines13 filesincludes scripts
Agents Manager
Manage all Clawdbot agents centrally: discover available agents, profile their capabilities, define routing hierarchy (who can assign to whom, who reports to whom), and intelligently route new tasks through the escalation chain.
🤖 For Agents
System Prompt (Copy-Paste)
Give this prompt to any agent to teach them this skill:
"You have access to the agents-manager skill. Use it to discover peers (scan_agents), check permissions (can_assign.js), and route tasks. Always follow the Handshake Protocol: Check requires_approval before assigning. If true, ask me first."
Core Rules
Discovery: Always check scan_agents.js before assuming an agent exists.
Permission: Never assign a task without checking can_assign.js or agent-registry.md.
protocol:
IF requires_approval is FALSE -> Assign directly.
IF requires_approval is TRUE -> Ask supervisor (Human or Agent).
👤 For Humans
Quick Start
Goal
Command
Setup
node scripts/setup_wizard.js (Run this first!)
List
node scripts/scan_agents.js
Health
node scripts/health_check.js
Stats
node scripts/log_analyzer.js
1. Agent Discovery & Profiling
List and profile all agents to understand their capabilities and routing configuration.
# List all agents
node {baseDir}/scripts/scan_agents.js
# Profile specific agent
node {baseDir}/scripts/generate_card.js <agent_id>