autonomous-agent-patterns
Design patterns for building autonomous coding agents. Covers tool integration, permission systems, browser automation, and human-in-the-loop workflows. Use when building AI agents, designing tool APIs, implementing permission systems, or creating autonomous coding assistants.
82
Quality
82%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Critical
Do not install without reviewing
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a well-crafted skill description that excels across all dimensions. It provides specific capabilities (tool integration, permission systems, browser automation, human-in-the-loop workflows), uses natural trigger terms users would actually say, and includes an explicit 'Use when...' clause with clear selection criteria. The description carves out a distinct niche for agent-building that won't conflict with general coding skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'tool integration, permission systems, browser automation, and human-in-the-loop workflows'. These are distinct, actionable capabilities rather than vague abstractions. | 3 / 3 |
Completeness | Clearly answers both what ('Design patterns for building autonomous coding agents. Covers tool integration, permission systems, browser automation, and human-in-the-loop workflows') and when ('Use when building AI agents, designing tool APIs, implementing permission systems, or creating autonomous coding assistants'). | 3 / 3 |
Trigger Term Quality | Includes natural keywords users would say: 'AI agents', 'tool APIs', 'permission systems', 'autonomous coding assistants', 'browser automation'. These cover common variations of how users might describe agent-building tasks. | 3 / 3 |
Distinctiveness Conflict Risk | Clear niche focused on autonomous agent design patterns with distinct triggers like 'AI agents', 'tool APIs', 'permission systems', and 'autonomous coding assistants'. Unlikely to conflict with general coding or documentation skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a comprehensive and highly actionable skill with excellent executable code examples covering agent architecture, tool design, permissions, and browser automation. The main weaknesses are its length (could benefit from splitting into multiple files) and the lack of explicit validation/recovery workflows for the complex multi-step processes it describes.
Suggestions
Split detailed implementations (browser automation, MCP integration, context management) into separate reference files, keeping SKILL.md as a concise overview with links
Add explicit validation checkpoints and error recovery workflows, especially for the sandboxed execution and tool creation patterns
Remove the 'When to Use This Skill' section as it duplicates the frontmatter description
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is comprehensive but includes some unnecessary verbosity, such as the 'When to Use This Skill' section that restates the description, and some code comments that explain obvious concepts. The ASCII diagram and extensive code examples add value but could be tightened. | 2 / 3 |
Actionability | Provides fully executable Python code examples throughout, including complete class implementations for AgentLoop, Tool schemas, EditFileTool, ApprovalManager, SandboxedExecution, and BrowserTool. Code is copy-paste ready with proper imports and error handling. | 3 / 3 |
Workflow Clarity | The agent loop diagram and code show clear sequencing, but the skill lacks explicit validation checkpoints for risky operations. The permission system is well-defined, but there's no clear workflow for 'if validation fails, do X' patterns when building agents. | 2 / 3 |
Progressive Disclosure | Content is well-organized with clear section headers and a logical progression from core architecture to advanced patterns. However, this is a monolithic 400+ line file that could benefit from splitting detailed implementations (browser automation, MCP integration) into separate reference files. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (762 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- duclm1x1/Dive-Ai
- Commit
20ba150
- Reviewed
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.