<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/how-i-deleted-95-of-my-agent-skills-and-got-better-results-nick-nisi-workos-qrj9q1mxo" -->

---
title: How I deleted 95% of my agent skills and got better...
description: A DX engineer at WorkOS shares lessons from building two AI agent systems: an internal multi-agent harness called 'Case' that automates PR creation across 20+...
canonical: https://daily.dev/posts/how-i-deleted-95-of-my-agent-skills-and-got-better-results-nick-nisi-workos-qrj9q1mxo
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: How I deleted 95% of my agent skills and got better results — Nick Nisi, WorkOS | daily.dev
og:description: A DX engineer at WorkOS shares lessons from building two AI agent systems: an internal multi-agent harness called 'Case' that automates PR creation across 20+...
og:url: https://daily.dev/posts/how-i-deleted-95-of-my-agent-skills-and-got-better-results-nick-nisi-workos-qrj9q1mxo
og:image: https://api.daily.dev/og/posts/QRj9Q1MXO.png
og:image:alt: How I deleted 95% of my agent skills and got better results — Nick Nisi, WorkOS
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# How I deleted 95% of my agent skills and got better results — Nick Nisi, WorkOS

**[AI Engineer](https://daily.dev/sources/aidotengineer)** · 17 min read · 1 upvotes · 0 comments

## Summary

A DX engineer at WorkOS shares lessons from building two AI agent systems: an internal multi-agent harness called 'Case' that automates PR creation across 20+ repos, and an external CLI that installs AuthKit into customer projects. Key insights include: enforce behavior through code (state machines) rather than prompts; replace trust with cryptographic proof that agents actually ran tests or fixed bugs; and counter-intuitively, deleting 95% of auto-generated skills (from 10,000 to 553 lines of targeted gotchas) improved agent performance from 77% to 97% accuracy on key tasks. Evals were critical to discovering that more context was actively hurting results. The retrospective agent pattern — where the system analyzes its own logs and updates memory files to avoid repeating mistakes — is highlighted as a key architectural decision.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.youtube.com/watch?v=vy7o1g2iHY8>

## Similar posts on daily.dev

- [Agentic Skill Decay](https://daily.dev/posts/agentic-skill-decay-nzqkcham8) · Addy Osmani · 89 upvotes · 13 comments
- [We Ran 250 AI Agent Evals to Find Out if Skills Beat Docs. The Answer Is More Complicated Than We Expected](https://daily.dev/posts/we-ran-250-ai-agent-evals-to-find-out-if-skills-beat-docs-the-answer-is-more-complicated-than-we-ex-uquqegqzg) · Wix Engineering · 0 upvotes · 0 comments
- [AGENTS.md outperforms skills in our agent evals](https://daily.dev/posts/agents-md-outperforms-skills-in-our-agent-evals-9b0bolfqt) · Hacker News · 20 upvotes · 0 comments

---

Tags: [#general-programming](https://daily.dev/tags/general-programming), [#ai-agents](https://daily.dev/tags/ai-agents), [#claude](https://daily.dev/tags/claude), [#prompt-engineering](https://daily.dev/tags/prompt-engineering)

[View this post on daily.dev](https://daily.dev/posts/how-i-deleted-95-of-my-agent-skills-and-got-better-results-nick-nisi-workos-qrj9q1mxo)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"How I deleted 95% of my agent skills and got better results — Nick Nisi, WorkOS","url":"https://daily.dev/posts/how-i-deleted-95-of-my-agent-skills-and-got-better-results-nick-nisi-workos-qrj9q1mxo","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/how-i-deleted-95-of-my-agent-skills-and-got-better-results-nick-nisi-workos-qrj9q1mxo"},"datePublished":"2026-05-30T18:00:18.557Z","dateModified":"2026-05-30T18:26:11.819Z","description":"A DX engineer at WorkOS shares lessons from building two AI agent systems: an internal multi-agent harness called 'Case' that automates PR creation across 20+...","image":"https://i.ytimg.com/vi/vy7o1g2iHY8/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/vy7o1g2iHY8/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AI Engineer","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"AI Engineer","logo":"https://media.daily.dev/image/upload/s--u5PucxNT--/f_auto/v1724338940/logos/aidotengineer","url":"https://daily.dev/sources/aidotengineer"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/how-i-deleted-95-of-my-agent-skills-and-got-better-results-nick-nisi-workos-qrj9q1mxo","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"general-programming,ai-agents,claude,prompt-engineering","timeRequired":"PT17M","video":{"@type":"VideoObject","name":"How I deleted 95% of my agent skills and got better results — Nick Nisi, WorkOS","description":"A DX engineer at WorkOS shares lessons from building two AI agent systems: an internal multi-agent harness called 'Case' that automates PR creation across 20+...","thumbnailUrl":"https://i.ytimg.com/vi/vy7o1g2iHY8/sddefault.jpg","uploadDate":"2026-05-30T18:00:18.557Z","duration":"PT17M","url":"https://api.daily.dev/r/QRj9Q1MXO","embedUrl":"https://www.youtube.com/embed/vy7o1g2iHY8"}}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AI Engineer","item":"https://daily.dev/sources/aidotengineer"},{"@type":"ListItem","position":3,"name":"How I deleted 95% of my agent skills and got better results — Nick Nisi, WorkOS"}]}
```

