autonomous-agent-patterns
Design patterns for building autonomous coding agents, inspired by [Cline](https://github.com/cline/cline) and [OpenAI Codex](https://github.com/openai/codex).
33
18%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Risky
Do not use without reviewing
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/autonomous-agent-patterns/SKILL.mdQuality
Discovery
22%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description is too abstract and lacks concrete actions, explicit trigger conditions, and natural user-facing keywords. It reads more like a topic label than a functional skill description. The references to Cline and Codex add some distinctiveness but don't compensate for the missing specificity and completeness.
Suggestions
Add a 'Use when...' clause specifying triggers like 'building an AI coding agent', 'implementing agentic tool-use loops', 'designing agent architectures', or 'creating autonomous code generation workflows'.
Replace the vague 'design patterns' with specific concrete actions such as 'Implements agent loops, tool-calling patterns, file-editing strategies, and error-recovery mechanisms for autonomous coding agents'.
Include natural trigger terms users would say, such as 'AI agent', 'agentic coding', 'code agent architecture', 'tool use patterns', 'agent loop design'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description mentions 'design patterns for building autonomous coding agents' which is abstract and vague. It does not list any concrete actions like 'generates agent loops', 'implements tool-use patterns', or 'creates file-editing workflows'. | 1 / 3 |
Completeness | The description partially addresses 'what' (design patterns for coding agents) but is vague about specifics, and completely lacks a 'when' clause or any explicit trigger guidance for when Claude should select this skill. | 1 / 3 |
Trigger Term Quality | Contains some relevant keywords like 'autonomous coding agents', 'Cline', 'OpenAI Codex', and 'design patterns', but misses common natural terms users might say such as 'AI agent', 'code agent', 'agentic workflow', 'tool use', or 'agent architecture'. | 2 / 3 |
Distinctiveness Conflict Risk | The mention of Cline and OpenAI Codex provides some distinctiveness, and 'autonomous coding agents' is a somewhat specific niche. However, 'design patterns' is broad enough to overlap with general software architecture or coding best practices skills. | 2 / 3 |
Total | 6 / 12 Passed |
Implementation
14%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is an extensive catalog of design patterns for autonomous agents, but it suffers from severe verbosity—most of the code demonstrates patterns Claude already knows how to implement from brief descriptions. The content lacks workflow sequencing (how to actually build an agent step-by-step) and is entirely monolithic with no progressive disclosure. It reads more like a tutorial blog post than an efficient skill file.
Suggestions
Reduce to a concise overview (<100 lines) with brief pattern descriptions and key design decisions, moving detailed code examples to separate referenced files (e.g., tools.md, permissions.md, browser.md)
Replace verbose code implementations with compact specifications—e.g., instead of a full EditFileTool class, describe the search/replace pattern with validation in 5-10 lines and let Claude generate the implementation
Add a clear sequential workflow for building an agent: 1) Define tools → 2) Implement agent loop → 3) Add permissions → 4) Test with validation → 5) Add sandboxing, with explicit checkpoints at each stage
Remove sections that explain concepts Claude already knows well (basic class inheritance, file I/O, subprocess usage) and focus only on non-obvious design decisions and gotchas specific to agent development
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~600+ lines. Includes extensive boilerplate code that Claude already knows how to write (subprocess wrappers, file I/O, basic class patterns). The multi-model architecture section explains obvious concepts like using different models for different tasks. Much of this is reference-level code that doesn't teach novel patterns—Claude can generate all of this from brief descriptions. | 1 / 3 |
Actionability | The code examples are fairly concrete and near-executable, but they rely on undefined imports/classes (ToolResult, playwright, html_to_markdown, mcp.Server) and are more illustrative patterns than copy-paste ready implementations. The code is closer to well-structured pseudocode than truly executable examples—missing imports, undefined helper functions, and incomplete class definitions throughout. | 2 / 3 |
Workflow Clarity | Despite being about building autonomous agents (a complex multi-step process), there are no clear sequential workflows for actually building an agent. The content presents isolated patterns/classes but never sequences them into a coherent build process. No validation checkpoints for testing the agent, no feedback loops for debugging, and the checklist at the end is generic rather than actionable. | 1 / 3 |
Progressive Disclosure | Monolithic wall of code with no references to external files for detailed content. All six major sections with extensive code examples are inlined into a single document. This would be a prime candidate for splitting into separate files (tools.md, permissions.md, browser.md, etc.) with a concise overview in the main skill file. | 1 / 3 |
Total | 5 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (765 lines); consider splitting into references/ and linking | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Repository
- sickn33/antigravity-awesome-skills
- Commit
d739c8b
- 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.