Generates Custom Agent files (.github/agents/*.agent.md) with persona-based configurations, specialized tool sets, and role-specific cognitive architectures. Use when the user requests a specialized role or perspective (Security Auditor, Documentation Writer, Release Engineer, etc.). Creates agents with defined identity, constrained tools, and thinking processes aligned to their role. Not for general capabilities—use generate-agent-skills for those.
93
—
Does it follow best practices?
Impact
90%
1.15xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Read-only analyst agent with role-specific cognitive architecture
Correct file path
100%
100%
Valid YAML frontmatter
100%
50%
Kebab-case agent name
100%
100%
Read-only tool set
30%
80%
XML delimiters present
0%
100%
Role-specific thinking process
40%
100%
Agent identity section
75%
100%
Role boundaries defined
62%
100%
Unique value articulated
87%
100%
Thinking process components
90%
100%
Tool justification present
87%
100%
agent-summary.md created
100%
100%
Writer agent with justified write access and explicit tool constraints
Correct file path
100%
100%
Write tools included
100%
37%
No execute-only tools for writer
100%
87%
File pattern constraints
100%
100%
XML delimiters present
0%
25%
Documentation-specific thinking
50%
100%
YAML frontmatter complete
100%
57%
Source code exclusion stated
100%
100%
Tool justification in decisions.md
100%
100%
Unique value articulated
100%
100%
Role constraints: NO code modification
100%
100%
Agent identity section
75%
100%
Release maintainer agent with execute tools, constraints, and single-role focus
Correct file path
100%
100%
Execute tools justified
80%
90%
Prohibited bash commands listed
100%
100%
Write tools included for release artifacts
100%
100%
XML delimiters present
0%
62%
Release-specific thinking process
50%
100%
YAML frontmatter complete
83%
50%
Single role, no mixing
100%
100%
Unique value over default agent
100%
100%
Block vs proceed criteria
100%
100%
Requires-approval constraint
100%
100%
Agent identity section
71%
100%
Semantic versioning mentioned
100%
100%
d1b6440
Table of Contents
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.