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. This skill covers agent loops (ReAct, Plan-Execute), goal decomposition, reflection patterns, and production reliability. Key insight: compounding error rates kill autonomous agents. A 95% success rate per step drops to 60% b
41
27%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.agent/skills/autonomous-agents/SKILL.mdQuality
Discovery
32%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 description reads more like an educational overview or article introduction than a skill description for Claude's skill selection. It explains concepts about autonomous agents but fails to specify what actions Claude performs or when to use this skill. The description also appears truncated, ending mid-sentence.
Suggestions
Add an explicit 'Use when...' clause with natural trigger phrases like 'building an agent', 'autonomous workflow', 'agent reliability', 'ReAct pattern'
Replace conceptual explanations with concrete actions Claude performs, e.g., 'Designs agent architectures, implements ReAct loops, debugs agent failures'
Complete the truncated description and remove editorial commentary ('The challenge isn't...', 'Key insight:') that doesn't help with skill selection
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (autonomous agents) and lists some concepts (agent loops, ReAct, Plan-Execute, goal decomposition, reflection patterns), but these are conceptual topics rather than concrete actions Claude would perform. No specific verbs describing what Claude does with this skill. | 2 / 3 |
Completeness | Provides conceptual 'what' (covers agent loops, patterns, reliability) but no explicit 'when' clause. The description reads more like educational content about agents than guidance for when Claude should select this skill. Also appears truncated mid-sentence. | 1 / 3 |
Trigger Term Quality | Includes some relevant technical terms like 'autonomous agents', 'ReAct', 'Plan-Execute', 'goal decomposition', but these are jargon-heavy. Missing natural user phrases like 'build an agent', 'create autonomous workflow', or 'agent reliability'. | 2 / 3 |
Distinctiveness Conflict Risk | The focus on autonomous agents and specific patterns like ReAct provides some distinctiveness, but the broad scope ('production reliability', 'goal decomposition') could overlap with general coding, architecture, or planning skills. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
22%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill content is incomplete (truncated mid-sentence) and functions more as a philosophical introduction with a table of contents than actionable guidance. While it correctly identifies the core challenge of compounding error rates in autonomous agents, it provides no concrete implementation details, code examples, or workflows in the main file. The sub-skill structure is reasonable but the parent skill needs substantive content of its own.
Suggestions
Add at least one concrete, executable code example showing a minimal agent loop implementation in the main skill file
Include a clear workflow showing how to build an agent incrementally: 1) single tool → 2) tool chain → 3) decision loop → 4) full autonomy, with validation at each stage
Provide a quick-start section with copy-paste ready code before linking to detailed sub-skills
Complete the truncated content and remove narrative padding ('You are an agent architect...') in favor of direct instruction
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The philosophical framing ('You are an agent architect who has learned...') adds some unnecessary narrative padding, but the core insight about compounding error rates is valuable and efficiently stated. The content is incomplete (cuts off mid-sentence) which affects assessment. | 2 / 3 |
Actionability | No concrete code, commands, or executable examples are provided. The content describes philosophy and links to sub-skills but offers no actionable guidance in the main file itself - just abstract principles like 'guardrails before capabilities.' | 1 / 3 |
Workflow Clarity | No workflow is defined. The skill mentions patterns (ReAct, Plan-Execute) but only as links to sub-skills. There are no steps, sequences, or validation checkpoints in the main content for implementing autonomous agents. | 1 / 3 |
Progressive Disclosure | The structure attempts progressive disclosure with links to sub-skills, which is good. However, the main file provides almost no substantive overview content - it's essentially just a table of contents with philosophy, making the references feel like the only content rather than supplements. | 2 / 3 |
Total | 6 / 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 |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- Dokhacgiakhoa/antigravity-ide
- Commit
332e58b
- 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.