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

55%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

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

Quality

Content

61%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 body is a well-executed pointer skill: concise, honest about limitations, and backed by a real, well-organized one-level-deep reference. Its weakness is that all actionable content is delegated — no quick-start guidance, key patterns, or section index appears in SKILL.md itself, so the skill offers nothing to act on without loading the full guide.

Suggestions

Add a short section index of the guide (e.g., '1. Core Agent Architecture · 2. Tool Design · 3. Permission & Safety · 4. Browser Automation · 5. Context Management · 6. MCP Integration') so 'load the relevant sections' is actionable without opening the whole file.

Inline 3-5 headline patterns or a minimal quick-start snippet (e.g., the agent-loop pseudocode or a tool schema skeleton) so the skill provides immediate value before the guide is read.

Trim the duplicated blockquote that repeats the frontmatter description verbatim, and tighten the generic Limitations boilerplate.

Dimension	Reasoning	Score
Conciseness	The body is lean (~25 lines) and offloads all substance to the reference, but two minor pieces could be trimmed: the opening blockquote duplicates the description verbatim, and the generic Limitations boilerplate ('Do not treat the output as a substitute for... expert review') adds little. Matches 'efficient; minor instances of over-explanation that could be trimmed' rather than the every-token-earns-its-place anchor at 5.	4 / 5
Actionability	The only concrete guidance is the pointer 'Read [the detailed guide](references/detailed-guide.md) before executing this skill' — a real, correct path, but the body contains no code, commands, or summary of the key patterns, so a reader must open the full 751-line guide for anything actionable. Fits 'some concrete guidance but incomplete'; not 4 because there is no executable substance in the body itself.	3 / 5
Workflow Clarity	The body establishes a rough sequence (read the guide — mandatory safety/prerequisites — then execute, choosing focused sections vs. end-to-end), but no step sequence or validation checkpoints appear in the body itself; the actual workflow lives entirely in the reference. Matches 'sequence present but checkpoints missing or implicit' better than the many-gaps anchor at 2, since the single navigation instruction is unambiguous.	3 / 5
Progressive Disclosure	The bundle structure is sound: one clearly signaled reference (references/detailed-guide.md) that exists, is well-sectioned (numbered sections 1-6 plus checklists), and is exactly one level deep with no nested references. It falls short of the 5 anchor because 'For focused work, load the relevant sections' is not supported by a section index in SKILL.md — selective loading requires opening the entire 22 KB guide.	4 / 5
	Total	14 / 20 Passed

Description

48%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 identifies a clear, distinct niche but under-delivers on specifics: it states one generic action, omits any 'when to use' trigger guidance, and lacks the natural keyword variations that would help it fire. Adding a Use-when clause and naming the concrete pattern areas (tool design, permissions, browser automation, context management) would lift it substantially.

Suggestions

Add an explicit trigger clause, e.g., 'Use when building autonomous AI/coding agents, designing tool-calling APIs, or implementing permission/approval systems for agents.'

Replace the generic 'Design patterns' phrasing with 2-3 concrete capability areas drawn from the guide (agent loop architecture, tool schema design, permission/sandboxing patterns, context management).

Include natural synonym terms users would say — 'AI agent', 'LLM agent', 'agentic coding', 'human-in-the-loop' — to improve trigger-term coverage.

Dimension	Reasoning	Score
Specificity	The description names the domain ("autonomous coding agents") but the only stated action is the generic "Design patterns for building" — no concrete capabilities (e.g., agent loops, tool schemas, permission systems, browser automation) appear, matching the 'names the domain but actions are minimal or generic' anchor rather than the 1-2-concrete-actions anchor at 3.	2 / 5
Completeness	It clearly answers 'what' (design patterns for building autonomous coding agents) but contains no 'Use when...' clause or equivalent trigger guidance in the description itself — the When-to-Use list lives only in the body — which caps completeness at 3 per the judging guidelines. Not 2 because the 'what' is clear rather than vague.	3 / 5
Trigger Term Quality	"autonomous coding agents" and the named tools "Cline"/"OpenAI Codex" are relevant keywords, but common variations users would actually say — "AI agent", "LLM agent", "agent loop", "tool calling", "human-in-the-loop" — are absent, fitting the 'some relevant keywords but missing common variations or synonyms' anchor. Not 4 because coverage is thin beyond the single domain phrase.	3 / 5
Distinctiveness Conflict Risk	"Autonomous coding agents" plus the specific Cline/Codex inspirations carve out a fairly distinct niche with minor overlap risk against closely related general agent-building skills, matching the 'mostly distinct' anchor. Not 5 because the description alone gives no distinct trigger phrases to disambiguate from other agent/tool-design skills.	4 / 5
	Total	12 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: boisenoise/skills-collections
Path: skills/antigravity-autonomous-agent-patterns/SKILL.md
Commit: 690de8f

Reviewed: 11 minutes 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.