--- title: "awesome-claude-skills vs agents-from-scratch" type: "comparison" canonical_url: "https://www.graphcanon.com/compare/composiohq-awesome-claude-skills-vs-pguso-agents-from-scratch" tools: ["composiohq-awesome-claude-skills", "pguso-agents-from-scratch"] --- # awesome-claude-skills vs agents-from-scratch *GraphCanon updated Jul 15, 2026* ## Verdict Pick awesome-claude-skills when pricing: The repository's license is under Apache 2.0 for the overall content, but individual skills may have varying licensing terms which should be checked individually within their respective folders.; pick agents-from-scratch when tags unique to agents-from-scratch: agent-architecture, ai-agents, ai-education, ai-from-scratch. [awesome-claude-skills](https://github.com/ComposioHQ/awesome-claude-skills) reports 67k GitHub stars, 7.6k forks, and 974 open issues, last pushed May 22, 2026. [agents-from-scratch](https://github.com/pguso/agents-from-scratch) has 901 stars, 226 forks, and 6 open issues, last pushed Jan 14, 2026. Figures are from public GitHub metadata via [awesome-claude-skills's repository](https://github.com/ComposioHQ/awesome-claude-skills) and [agents-from-scratch's repository](https://github.com/pguso/agents-from-scratch). | | [awesome-claude-skills](/tools/composiohq-awesome-claude-skills.md) | [agents-from-scratch](/tools/pguso-agents-from-scratch.md) | | --- | --- | --- | | Tagline | A curated list of awesome Claude Skills for customizing AI workflows | Build AI agents from first principles using a local LLM - no frameworks, no cloud APIs, no hidden reasoning. | | Stars | 67,447 | 901 | | Forks | 7,586 | 226 | | Open issues | 974 | 6 | | Language | Python | Python | | Adopt for | awesome-claude-skills is a curated list that provides resources and tools for customizing workflows using Claude AI. | - | | Persona | - | - | | Runtime | - | - | | License | - | MIT | | Categories | AI Agents, Developer Tools | AI Agents, LLM Frameworks | ## Trust and health _Sourced signals - not a safety guarantee. No winner column._ | | [awesome-claude-skills](/tools/composiohq-awesome-claude-skills.md) | [agents-from-scratch](/tools/pguso-agents-from-scratch.md) | | --- | --- | --- | | Maintenance | Steady (60%) | Slowing (36%) | | Days since push | 50d | 182d | | Open issues (now) | 974 | 6 | | Owner type | Organization | User | | Full report | [trust report](/tools/composiohq-awesome-claude-skills/trust.md) | [trust report](/tools/pguso-agents-from-scratch/trust.md) | ## Decision facts: awesome-claude-skills - **Pricing:** unknown - The repository's license is under Apache 2.0 for the overall content, but individual skills may have varying licensing terms which should be checked individually within their respective folders. - **Adopt for:** awesome-claude-skills is a curated list that provides resources and tools for customizing workflows using Claude AI. ## Choose when ### Choose awesome-claude-skills if… - Pricing: The repository's license is under Apache 2.0 for the overall content, but individual skills may have varying licensing terms which should be checked individually within their respective folders.. - Tags unique to awesome-claude-skills: agent-skills, automation, claude code, composio. - Also covers Developer Tools. - When you are looking to customize and extend the capabilities of Claude AI through various skills and plugins. ### Choose agents-from-scratch if… - Tags unique to agents-from-scratch: agent-architecture, ai-agents, ai-education, ai-from-scratch. - Also covers LLM Frameworks. - Leaner open-issue backlog (6). ## When NOT to use awesome-claude-skills - Avoid using if your primary focus is on general-purpose automation that doesn't require integration with Claude AI or its ecosystem. - Not recommended if specific automation tasks do not benefit from customization through the provided Claude Skills, or you prefer tools that operate independently without dependency on a particular AI. ## When NOT to use agents-from-scratch - Last GitHub push was 182 days ago (slowing maintenance, Jan 14, 2026). Validate activity before betting a new project on agents-from-scratch. - AI Agents: Don't use an agent loop when a deterministic workflow would do; agents add latency, cost, and non-determinism. - LLM Frameworks: Avoid a framework for a single prompt-and-retrieve call; the abstraction can cost more than it saves. ## Common questions ### What is the difference between awesome-claude-skills and agents-from-scratch? awesome-claude-skills: A curated list of awesome Claude Skills for customizing AI workflows. agents-from-scratch: Build AI agents from first principles using a local LLM - no frameworks, no cloud APIs, no hidden reasoning.. See the comparison table for live GitHub stats and shared categories. ### When should I choose awesome-claude-skills over agents-from-scratch? Choose awesome-claude-skills over agents-from-scratch when Pricing: The repository's license is under Apache 2.0 for the overall content, but individual skills may have varying licensing terms which should be checked individually within their respective folders.; Tags unique to awesome-claude-skills: agent-skills, automation, claude code, composio; Also covers Developer Tools; When you are looking to customize and extend the capabilities of Claude AI through various skills and plugins. ### When should I choose agents-from-scratch over awesome-claude-skills? Choose agents-from-scratch over awesome-claude-skills when Tags unique to agents-from-scratch: agent-architecture, ai-agents, ai-education, ai-from-scratch; Also covers LLM Frameworks; Leaner open-issue backlog (6). ### When should I avoid awesome-claude-skills? Avoid using if your primary focus is on general-purpose automation that doesn't require integration with Claude AI or its ecosystem. Not recommended if specific automation tasks do not benefit from customization through the provided Claude Skills, or you prefer tools that operate independently without dependency on a particular AI. ### When should I avoid agents-from-scratch? Last GitHub push was 182 days ago (slowing maintenance, Jan 14, 2026). Validate activity before betting a new project on agents-from-scratch. AI Agents: Don't use an agent loop when a deterministic workflow would do; agents add latency, cost, and non-determinism. LLM Frameworks: Avoid a framework for a single prompt-and-retrieve call; the abstraction can cost more than it saves. ### Is awesome-claude-skills or agents-from-scratch more popular on GitHub? awesome-claude-skills has more GitHub stars (67,447 vs 901). Stars measure visibility, not whether either tool fits your constraints. ### Are awesome-claude-skills and agents-from-scratch open source? Yes - both are open-source projects on GitHub. ### Where can I find alternatives to awesome-claude-skills or agents-from-scratch? GraphCanon lists graph-backed alternatives at [awesome-claude-skills alternatives](/tools/composiohq-awesome-claude-skills/alternatives) and [agents-from-scratch alternatives](/tools/pguso-agents-from-scratch/alternatives) ([awesome-claude-skills markdown twin](/tools/composiohq-awesome-claude-skills/alternatives.md), [agents-from-scratch markdown twin](/tools/pguso-agents-from-scratch/alternatives.md)), ranked by typed relationship edges rather than popularity votes. ### Is there a machine-readable version of this comparison? Yes. The markdown twin at [this comparison](/compare/composiohq-awesome-claude-skills-vs-pguso-agents-from-scratch.md) mirrors this page for agents and LLM crawlers, with the same stats table and FAQ answers. ### Which is better maintained, awesome-claude-skills or agents-from-scratch? awesome-claude-skills: Steady. agents-from-scratch: Slowing. Compare maintenance labels, days since push, and release cadence in the trust section below - stars alone do not measure maintenance. ### Where are the full trust reports for awesome-claude-skills and agents-from-scratch? GraphCanon publishes per-repo trust reports with dated maintenance, provenance, and scan summaries: [awesome-claude-skills trust report](/tools/composiohq-awesome-claude-skills/trust); [agents-from-scratch trust report](/tools/pguso-agents-from-scratch/trust). --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/graph?tool=composiohq-awesome-claude-skills`](/api/graphcanon/graph?tool=composiohq-awesome-claude-skills) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_