agent-factory
Claude Code agent generation system that creates custom agents and sub-agents with enhanced YAML frontmatter, tool access patterns, and MCP integration support following proven production patterns
Install with Tessl CLI
npx tessl i github:alirezarezvani/claude-code-skill-factory --skill agent-factory57
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillValidation for skill structure
Evaluation results
100%
32%Security Audit Agent for E-Commerce Platform
Quality agent type configuration
Kebab-case name
100%
100%
Tools comma-separated
0%
100%
Red color assigned
100%
100%
Quality tools included
0%
100%
Sonnet model
0%
100%
Sequential execution documented
100%
100%
Required YAML fields present
50%
100%
Specific description
100%
100%
Valid field category
100%
100%
Valid expertise level
100%
100%
Agent file path correct
100%
100%
Metadata output present
83%
100%
Without context: $0.4343 · 1m 59s · 27 turns · 144 in / 6,553 out tokens
With context: $0.6889 · 2m 2s · 28 turns · 3,201 in / 6,678 out tokens
92%
43%ML Project Planning Agent for Research Team
Strategic agent model and tool constraints
Opus model assigned
66%
100%
Blue color assigned
100%
100%
Strategic tool set only
0%
100%
Tools comma-separated
0%
100%
Kebab-case name
100%
100%
Required YAML fields
70%
100%
Valid strategic field
0%
0%
Valid expertise level
0%
100%
Specific description
87%
100%
Parallel-safe pattern noted
25%
100%
Validation report produced
50%
100%
Agent spec includes metadata
100%
100%
Without context: $0.2396 · 1m 13s · 14 turns · 16 in / 3,859 out tokens
With context: $0.5017 · 1m 27s · 22 turns · 1,287 in / 4,871 out tokens
95%
41%Agent Suite for a FinTech Data Pipeline
Multi-agent workflow with correct type assignments
Coordination agent purple
100%
100%
Coordination agent limited tools
0%
100%
Implementation agent green
0%
100%
Implementation agent full tools
100%
100%
Domain specialist orange
0%
100%
Tools comma-separated all agents
0%
100%
Kebab-case all agent names
0%
100%
Focused single responsibilities
100%
100%
Correct model assignments
100%
50%
Execution patterns documented
70%
100%
Required YAML fields all agents
62%
100%
Suite summary produced
100%
100%
Without context: $0.5279 · 2m 25s · 29 turns · 961 in / 8,283 out tokens
With context: $0.8821 · 2m 56s · 33 turns · 799 in / 9,537 out tokens
- Evaluated
- Agent
- Claude Code
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.