autonomous-agents

Autonomous agents are AI systems that can independently decompose goals, plan actions, execute tools, and self-correct without constant human guidance. The challenge isn't making them capable - it's making them reliable. Every extra decision multiplies failure probability.

Quality

50%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/autonomous-agents/SKILL.md

Quality

Content

46%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 rich with concrete, mostly-executable code and a useful validation checklist, but it is a monolithic 1085-line reference that explains general agent-failure concepts Claude already knows and inlines material that belongs in separate reference files. Workflow clarity is bounded by the catalog structure and the destructive-operation cap.

Suggestions

Move the 'Sharp Edges' catalog and the larger code libraries into reference files under references/, leaving SKILL.md as a concise overview with one-level-deep links.

Cut the conceptual prose that re-explains known ideas (compounding error math, demo-vs-prod gap, fabrication tendencies) and keep only the non-obvious guidance.

Add an explicit sequenced build workflow (e.g. choose pattern -> add guardrails -> add checkpointing -> validate against the checklist) with validation checkpoints to raise workflow clarity above the destructive-op cap.

Dimension	Reasoning	Score
Conciseness	At ~1085 lines the body is noticeably verbose: it re-derives concepts Claude already knows (compounding error rates, demo-vs-production gap, why agents fabricate data) and repeats the description's thesis verbatim, with several padded prose sections alongside the useful code.	2 / 5
Actionability	Most sections give concrete, mostly executable Python (ReAct/LangGraph setup, GuardedAgent, CostLimitedAgent, checkpointing, retry clients) with specific config flags like max_iterations and interrupt_before; minor gaps come from undefined helper classes and a few approximate APIs.	4 / 5
Workflow Clarity	The skill is organized as a pattern catalog plus a 'Sharp Edges' catalog plus a 'Validation Checks' checklist rather than a single end-to-end build workflow, and because it covers destructive/batch agent operations the rubric caps workflow_clarity at 3 even though validation guidance is present.	3 / 5
Progressive Disclosure	No bundle files exist (references/scripts/assets are absent) and the entire 1085-line skill is inlined in SKILL.md; the large Sharp Edges catalog and code libraries clearly belong in separate reference files, so structure is minimal despite clear section headers.	2 / 5
	Total	11 / 20 Passed

Description

55%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 conveys a clear domain and a memorable reliability thesis, but it reads as a definition plus commentary rather than a capability+trigger statement, and it lacks an explicit 'Use when...' clause. Trigger-term coverage is thin on natural user phrasings.

Suggestions

Add an explicit 'Use when...' clause naming natural trigger phrases (e.g. 'Use when the user mentions autonomous agents, agentic AI, agent loops, ReAct, or goal decomposition').

Reframe the opening as skill capabilities in third person (e.g. 'Designs and hardens autonomous agent loops...') instead of a dictionary definition.

Include common synonyms users actually say ('agentic AI', 'AI agents', 'LLM agents') to improve trigger-term coverage and distinctiveness.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('decompose goals, plan actions, execute tools, and self-correct'), but the framing is definitional ('Autonomous agents are AI systems that can...') rather than direct skill capabilities, and the trailing commentary is not action language.	4 / 5
Completeness	It gives a clear 'what' (defines the agent domain and the reliability focus) but includes no 'Use when...' clause or equivalent explicit trigger guidance, so per the rubric completeness is capped at 3.	3 / 5
Trigger Term Quality	The main natural keyword is 'autonomous agents'; the remaining terms (decompose goals, self-correct, failure probability) are descriptive rather than phrases users naturally say, and common synonyms like 'agentic', 'AI agent', or 'LLM agent' are absent.	3 / 5
Distinctiveness Conflict Risk	'Autonomous agents' is a recognizable niche, but the term is broad within the agent ecosystem and overlaps with closely related skills (multi-agent-orchestration, agent-memory-systems, agent-tool-builder) that the body itself lists.	3 / 5
	Total	13 / 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 (1085 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-agents/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.