tessl install https://github.com/jezweb/claude-skills --skill agent-developmentgithub.com/jezweb/claude-skills
Design and build custom Claude Code agents with effective descriptions, tool access patterns, and self-documenting prompts. Covers Task tool delegation, model selection, memory limits, and declarative instruction design. Use when: creating custom agents, designing agent descriptions for auto-delegation, troubleshooting agent memory issues, or building agent pipelines.
Average Score
96%
Content
92%
Description
100%
Generated
Validations
Total score
15/16| Criteria | Score |
|---|---|
skill_md_line_count SKILL.md line count is 332 (<= 500) | |
frontmatter_valid YAML frontmatter is valid | |
name_field 'name' field is valid: 'agent-development' | |
description_field 'description' field is valid (372 chars) | |
description_voice 'description' uses third person voice | |
description_trigger_hint Description includes an explicit trigger hint | |
compatibility_field 'compatibility' field not present (optional) | |
allowed_tools_field 'allowed-tools' field not present (optional) | |
metadata_version 'metadata' field is not a dictionary | |
metadata_field 'metadata' field not present (optional) | |
license_field 'license' field is present: MIT | |
frontmatter_unknown_keys No unknown frontmatter keys found | |
body_present SKILL.md body is present | |
body_examples Examples detected (code fence or 'Example' wording) | |
body_output_format Output/return/format terms detected | |
body_steps Step-by-step structure detected (ordered list) |
Content
Suggestions 2
Total score
11/12| Dimension | Score |
|---|---|
conciseness The content is lean and efficient throughout. It uses tables for quick comparison, avoids explaining concepts Claude already knows (like what agents are), and every section provides actionable information without padding. | 3/3 |
actionability Provides fully executable code examples (YAML frontmatter, JSON settings, bash commands), specific patterns with copy-paste ready templates, and concrete comparison tables showing exactly what to do vs. what not to do. | 3/3 |
workflow_clarity Multi-step processes are clearly sequenced with explicit validation steps. The workflow pattern section shows numbered steps with clear roles (ME vs AGENTS), and the pipeline agents section includes verification commands to catch common bugs. | 3/3 |
progressive_disclosure Content is well-organized with clear sections and headers, but it's a long monolithic document (~300 lines) that could benefit from splitting advanced topics (pipeline agents, parallel workflows) into separate reference files with links from the main skill. | 2/3 |
Suggestions
Consider splitting 'Pipeline Agents', 'Workflow Pattern', and 'Signs a Task Fits This Pattern' into a separate ADVANCED_PATTERNS.md file with a link from the main skill
Add a brief 'Quick Start' section at the top with just the essential agent frontmatter template before diving into detailed patterns
Overall Assessment
This is a high-quality, comprehensive skill for agent development that excels at actionability and conciseness. The content provides concrete patterns, executable templates, and clear anti-patterns throughout. The main weakness is that the document is quite long and could benefit from splitting advanced topics into separate files for better progressive disclosure.
Description
Total score
12/12| Dimension | Score |
|---|---|
specificity Lists multiple specific concrete actions: 'Design and build custom Claude Code agents', 'effective descriptions', 'tool access patterns', 'self-documenting prompts', 'Task tool delegation', 'model selection', 'memory limits', and 'declarative instruction design'. | 3/3 |
completeness Clearly answers both what (design/build agents with descriptions, tool access, prompts, delegation, model selection, memory limits) AND when (explicit 'Use when:' clause with four specific trigger scenarios). | 3/3 |
trigger_term_quality Includes natural keywords users would say: 'custom agents', 'agent descriptions', 'auto-delegation', 'agent memory issues', 'agent pipelines', 'Task tool', 'model selection'. These cover common variations of how users would describe agent-building tasks. | 3/3 |
distinctiveness_conflict_risk Clear niche focused specifically on Claude Code agent creation with distinct triggers like 'auto-delegation', 'agent pipelines', and 'agent memory issues' that are unlikely to conflict with general coding or documentation skills. | 3/3 |
Overall Assessment
This is a well-crafted skill description that excels across all dimensions. It provides specific capabilities, includes natural trigger terms users would actually say, explicitly separates 'what' from 'when', and carves out a distinct niche around Claude Code agent development that won't conflict with other skills.