# Skills are new features: Building Skill-Centric Harness — Yogendra Miraje, FactSet

Source: [AI Engineer](https://www.youtube.com/watch?v=7jjudsEhBtM)  
Feed7 permalink: https://feed7.dev/p/skills-are-new-features-building-skill-centric-harness-yogendra-miraje-f-0lp4c7o  
Published: 2026-07-29T18:00:06.000Z  
Trust: Source Linked (source_linked)

## Why Included

FactSet treats skills as versioned product features and the harness as their runtime. Routing descriptions, model-specific evals, access controls, and governance matter as libraries grow.

## Source Summary

A minimal skill-capable harness needs **a registry, a system prompt, and file reading**; executable skills also need a sandbox. The registry exposes names, descriptions, and paths, while the agent progressively loads only the instructions needed for a task.

## Practical Implication

Write descriptions as distinct routing signals tied to user intent, and organize skills around jobs rather than internal data models. Rerun evals after model changes because skills are **contracts versioned to a model**, not passive documentation.

## Agent-Ready Context

A minimal skill-capable harness needs **a registry, a system prompt, and file reading**; executable skills also need a sandbox. The registry exposes names, descriptions, and paths, while the agent progressively loads only the instructions needed for a task.

Write descriptions as distinct routing signals tied to user intent, and organize skills around jobs rather than internal data models. Rerun evals after model changes because skills are **contracts versioned to a model**, not passive documentation.

Putting every skill in the prompt stops scaling. The talk suggests shortlisting beyond roughly **10 skills** and adding hierarchy, metadata, ownership, lifecycle rules, audits, and tool allowlists when a library reaches hundreds.

## Connected Context

Feed7 judgment across 297 accumulated Signals:

This turns skills from Markdown conveniences into model-bound product contracts with routing, evaluation, ownership, security, and lifecycle requirements. It confirms progressive loading as the scaling mechanism, while narrowing simple skill-pack approaches: once libraries grow, descriptions must support reliable selection and executable skills require sandboxing, allowlists, audits, and reevaluation after model changes.

- [affaan-m/ECC](https://feed7.dev/p/ecc-0438yju) — ECC’s 261-skill breadth makes the talk’s hierarchy, shortlisting, selective installation, and duplicate-control requirements concrete rather than optional.
- [addyosmani/agent-skills](https://feed7.dev/p/agent-skills-0d3lx1j) — The portable Markdown pack exemplifies the minimal registry-and-file-loading model, while the talk adds that portability does not remove model-specific evaluation obligations.
- [We Vetted 2000 AI Skills Before They Reached Developers — Lucas Palma, Nubank](https://feed7.dev/p/we-vetted-2000-ai-skills-before-they-reached-developers-lucas-palma-nuba-0k4ehkz) — Nubank’s scanning and marketplace gates implement the supply-chain audits and lifecycle controls required when executable skill libraries expand.
- [WTF Is the Context Layer? The Missing Infrastructure for Production Agents — Prukalpa Sankar](https://feed7.dev/p/wtf-is-the-context-layer-the-missing-infrastructure-for-production-agent-0t47xqf) — A shared, versioned context layer offers infrastructure for the metadata and portability this skill-centric harness requires across changing agents.

## Context Map

- Layer: agent
- Domains: coding
- Topics: skills, harness-engineering, context-engineering

## Uncertainty

- Putting every skill in the prompt stops scaling. The talk suggests shortlisting beyond roughly **10 skills** and adding hierarchy, metadata, ownership, lifecycle rules, audits, and tool allowlists when a library reaches hundreds.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.