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).

Quality

47%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/autonomous-agent-patterns/SKILL.md

Quality

Content

53%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The skill is a rich, actionable catalog of executable agent-design patterns, but it is overlong, lacks sequenced validation workflows for its destructive-operation topics, and does not use progressive disclosure to split content into reference files.

Suggestions

Split each major pattern group (tool design, permissions/sandboxing, browser automation, context management, MCP) into separate reference files and keep SKILL.md as a concise overview with one-level-deep links.

Add explicit validation/feedback-loop steps to the permission and sandboxing sections (e.g., validate path -> attempt -> on failure report and halt) since they involve destructive operations.

Trim the ASCII agent-loop diagram and redundant boilerplate to reduce token cost while preserving the executable code examples.

Dimension	Reasoning	Score
Conciseness	The body is mostly code with little prose padding, but at ~760 lines it includes decorative elements like the ASCII agent-loop diagram and boilerplate that could be tightened; it is efficient in places yet overlong overall.	3 / 5
Actionability	It provides numerous concrete, largely executable Python classes (AgentLoop, EditFileTool, ApprovalManager, SandboxedExecution, BrowserTool) that are copy-paste ready, with only minor gaps such as undefined ToolResult/Enum imports.	4 / 5
Workflow Clarity	As a patterns catalog it has no single sequenced workflow, and where it touches destructive operations (sandboxing, permissions) it presents patterns rather than explicit validate->fix->retry checkpoints, capping it at 3.	3 / 5
Progressive Disclosure	No bundle files exist and the entire body is a monolithic single file; substantial per-pattern content that clearly belongs in separate reference files is inlined, despite clear section headers.	2 / 5
	Total	12 / 20 Passed

Description

41%Weight 40%Scale 1-5

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 states a clear purpose but is thin on concrete capabilities and entirely missing trigger guidance, so it reads more as a tagline than an activatable skill description.

Suggestions

List 2-4 concrete capabilities (e.g., 'design agent loops, tool schemas, permission systems, and context management') instead of the single generic action 'building'.

Add an explicit 'Use when...' clause naming natural trigger phrases users would say, such as 'Use when building autonomous AI agents, tool/function-calling APIs, or human-in-the-loop approval workflows'.

Include common synonyms and related terms (e.g., 'AI agents', 'tool calling', 'function calling', 'approval systems') to improve trigger-term coverage.

Dimension	Reasoning	Score
Specificity	The phrase "Design patterns for building autonomous coding agents" names the domain but offers only the single generic action of "building"; it lists no concrete capabilities, unlike the score-3 anchor which names 1-2 specific actions.	2 / 5
Completeness	It gives a clear "what" (design patterns for building autonomous coding agents) but provides no "when" guidance at all, so per the missing-trigger cap it cannot exceed 3.	3 / 5
Trigger Term Quality	It relies on the technical phrase "autonomous coding agents" plus project names (Cline, Codex) but lacks the natural phrases a user would actually say when needing this skill, with no "Use when..." trigger language.	2 / 5
Distinctiveness Conflict Risk	"Autonomous coding agents" is a fairly specific niche with distinct triggers and only minor overlap risk against general coding or prompting skills, fitting the mostly-distinct anchor better than the somewhat-broad score 3.	4 / 5
	Total	11 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (770 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: sickn33/antigravity-awesome-skills
Path: skills/autonomous-agent-patterns/SKILL.md
Commit: 1f67c44

Reviewed: about 17 hours ago

Table of Contents

Discovery Implementation Validation

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.