Urgent.News

What's breaking now, across thousands of outlets.

Tech

63,336 agent skills, 56 paid: what the pricing data actually says

If you have been thinking about selling a SKILL.md , this is the number that should stop you for a minute: in a frozen census of 63,336 agent skills taken on 2026-10-16, exactly 56 skills carry a paid label . That is 1.1% of the 4,902 skills that carry any pricing label at all. Source: Skillselion Agent Skill Pricing Census , edition of 2026-10-16, dataset frozen and downloadable under CC BY 4.0.…

On 2026-10-16, a census revealed that only 1.1% of SKILL.md agent skills carry a paid label, despite 4,902 skills having some pricing label. Of these priced skills, 70.4% are free to run, 24.8% are free with a BYO API key, and 3.7% are freemium. Paid skills only account for 0.5% of installs, with a total of 454,127 installs spread across 56 skills. Most of the costs in this market are due to separate paid services rather than the creator's price.

The issue isn't a pricing problem but rather a lack of market demand. SKILL.md is a text file that can be easily inspected, copied, and rewritten, and buyers cannot claim ownership of the service the skill drives. The data shows that payments move towards BYO-key skills, where the file is the on-ramp and the service behind it is the product. Install counts do not equate to usage, and there's a low first-run conversion rate, indicating that many users never actually engage with the skill.

To survive in this market, three models have proven successful: 1) The skill acts as an on-ramp to a paid service, 2) A curated bundle with a maintenance promise, and 3) The skill serves as an audit that provides a diagnostic, with the fix being the actual product. Priced like a book, a single SKILL.md has little chance of success, accounting for only 0.5% of installs and 1.1% of the market.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in Tech

Source Deduplication Patterns for a Healthcare Appointment Retrieval Architecture

Short answer: hash and version sources at ingest, group near-duplicate passages at retrieval, and require a current source ID plus page citation for every appointment claim.

  • Deduplication conducted in two stages: ingest and retrieval
  • Ingest stage uses hashes and similarity checks for repeated uploads and near-duplicates
  • Retrieval groups candidates by canonical source and selects best-scoring passage

The five-minute test that would have saved me eleven hours of directory submissions

Eleven hours of directory submissions produced two followed links. Not because the directories were fake, but because I never asked the one question that decides the outcome, and it takes five minutes…

  • Author discovers five-minute test to save eleven hours of directory submissions
  • Checks three details on listing page for linking potential
  • Single script used twice for both current and future submissions

Wabi as a Health and Habits Companion: 5 Trackers You Can Build Inside a Chat

I run Wabi Wiki, an unofficial guide to the Wabi AI assistant — not affiliated with Wabi. Most people meet Wabi as a chat app that builds mini-apps.

  • Five health trackers can be built within Wabi AI's chat-based mini-apps
  • Weight tracker logs weigh-ins and displays weight trend over time
  • Habit tracker sends gentle nudges and tracks streaks for daily habits

Day 26: Numeric Semantics, Multiple Currencies and Units of Measure

The chapters so far asked whether a value gets in, whether it gets across, and whether it gets recorded. This post asks a different question: what the value that got in actually stands for.

  • Numerical fields have decimal place limits not explicitly defined
  • Quantity, amount, and percentage fields are rounded
  • Exchange rates fixed at five decimal places

More from Thursday 19 November →