---
schema_version: "newruntime-topic-hub-v0.1"
type: "topic_hub"
slug: "agent-skills"
title: "Agent Skills - New Runtime"
description: "A New Runtime topic hub collecting signals, patterns, field notes, and public sources about agent skills."
answer: ["Agent Skills is tracked here as an evidence-linked topic, not as a static glossary entry.","The page connects raw observations to pattern hypotheses, longer analysis, and public sources.","Use it as the canonical landing page before drilling into individual records."]
search_intents: ["agent skills","agent skills AI agents","agent skills software"]
status: "tracked"
last_updated: "2026-07-16"
counts: {"total":5,"signals":4,"patterns":1,"posts":0,"atlas":0,"sources":10}
routes: {"html":"https://newruntime.com/topics/agent-skills/","markdown":"https://newruntime.com/topics/agent-skills.md","json":"https://newruntime.com/topics/agent-skills.json"}
top_sources: ["https://arxiv.org/abs/2602.12670","https://arxiv.org/abs/2603.29919","https://arxiv.org/abs/2605.24050","https://github.com/benchflow-ai/skillsbench","https://github.com/pydantic/pydantic-ai","https://github.com/shadcn/improve","https://github.com/shadcn/improve/blob/main/skills/improve/SKILL.md","https://platform.claude.com/docs/en/agents-and-tools/agent-skills/best-practices","https://pydantic.dev/articles/pydantic-ai-capabilities","https://youtube.com/watch?v=A8mokin_YOs"]
---

# Agent Skills - New Runtime

A New Runtime topic hub collecting signals, patterns, field notes, and public sources about agent skills.

## Short Answer

- Agent Skills is tracked here as an evidence-linked topic, not as a static glossary entry.
- The page connects raw observations to pattern hypotheses, longer analysis, and public sources.
- Use it as the canonical landing page before drilling into individual records.

## Patterns

- [Skills become a portable capability layer](https://newruntime.com/patterns/skills-become-portable-capability-layer/): Agent skills are emerging as a portable capability layer, but their value depends on progressive disclosure, provenance, security review, and behavioral evaluation.

## Field Notes


## Recent Raw Signals

- 2026-07-16: [Agent skills need behavioral evals, not prose review](https://newruntime.com/signals/agent-skills-need-behavioral-evals/)
- 2026-07-12: [Improve routes architecture and execution to different models](https://newruntime.com/signals/improve-routes-architect-and-executor-models/)
- 2026-07-12: [Large-task planning moves from chat into a ticket graph](https://newruntime.com/signals/large-task-planning-becomes-ticket-graph/)
- 2026-06-28: [Pydantic AI loads capabilities only when needed](https://newruntime.com/signals/pydantic-ai-capabilities-progressive-disclosure/)

## Public Sources

- https://arxiv.org/abs/2602.12670
- https://arxiv.org/abs/2603.29919
- https://arxiv.org/abs/2605.24050
- https://github.com/benchflow-ai/skillsbench
- https://github.com/pydantic/pydantic-ai
- https://github.com/shadcn/improve
- https://github.com/shadcn/improve/blob/main/skills/improve/SKILL.md
- https://platform.claude.com/docs/en/agents-and-tools/agent-skills/best-practices
- https://pydantic.dev/articles/pydantic-ai-capabilities
- https://youtube.com/watch?v=A8mokin_YOs