04-19-2026Price:

The Frontier

Your signal. Your price.

AI & TECH

Anthropic's modular skills fix AI decay

Sunday, April 19, 2026 · from 4 podcasts
  • Anthropic’s new modular skills prevent AI agents from breaking under bloated prompts.
  • Prompts are becoming reusable software libraries, not disposable chat text.
  • Reliability now trumps capability as the key metric for AI agents.

The era of dumping 5,000 tokens of instructions into an AI at once is over. Nathaniel Whittemore on The AI Daily Brief calls it 'context bloat' - a 2025 hangover where agents slowed to a crawl, then failed, under the weight of their own instructions. The fix isn’t better prompting. It’s software engineering.

Anthropic’s Claude Code team pioneered a shift to modular 'Agent Skills' using progressive disclosure. Instead of loading every rule upfront, agents now parse lightweight metadata and pull in specific skills - markdown files, scripts, or assets - only when needed. This keeps context clean and execution focused. OpenAI and GitHub Copilot have already adopted similar architectures.

"The era of the massive system prompt is over."

- Nathaniel Whittemore, The AI Daily Brief

Reliability is now the bottleneck. Whittemore notes that model updates routinely break old prompts, making maintenance unsustainable. Anthropic’s new Skill Creator tool addresses this by automatically rewriting vague skill descriptions, improving triggering accuracy in five out of six tests. It also enables A/B testing and benchmarking - treating skills like software, not text.

The mental model is shifting from one-off conversations to libraries of repeatable capabilities. Notion now lets users turn any page into a reusable skill. Replit’s Amjad Masad sees the same trend: AI agents now perform at Google mid-level engineer level, but only if they don’t get lost in their own instructions. The advantage now belongs to those who manage agents like codebases.

"Building true wealth requires prioritizing equity over high-salary 'bullshit work.'"

- Amjad Masad, The a16z Show

Prompts are no longer disposable. They’re durable assets. Whether it’s a custom workflow in Notion or a verification skill that checks an agent’s output, the goal is to teach the AI once and reuse it everywhere. The best skills encode human preferences - team processes, tone, decision logic - not just technical functions. These don’t expire when models improve.

Source Intelligence

- Deep dive into what was said in the episodes

Agent Building Trends [Operator Bonus Episode]Apr 18

Also from this episode: (15)

Other (15)

  • The agentic era of AI relies heavily on 'skills,' which are open-format folders containing instructions, scripts, and resources that agents can dynamically discover and use to enhance capabilities for specific tasks.
  • Agent skills emerged to solve the problem of continuously ballooning system prompts, which led to performance degradation, increased cost, and reduced reliability in AI coding agents by 2025.
  • Anthropic announced agent skills on October 16th, framing them as a way to equip models like Claude with procedural knowledge and organizational context to perform complex tasks using local code execution and file systems.
  • A skill is structured as a directory anchored by a `skill.md` file, employing progressive disclosure by initially presenting only metadata (name, short description) to the agent before revealing full content or linked resources.
  • Tariq of Anthropic's Claude Code team highlights that skills are not merely markdown files but folders capable of bundling scripts, assets, and data, allowing agents to discover, explore, and manipulate additional context.
  • Agent skills were rapidly adopted by various AI ecosystems, including OpenAI's ChatGPT and GitHub Copilot, leading to platforms like ClaudeHub amassing over 28,000 skills for common agent needs.
  • Anthropic identified nine common categories for most agent skills, ranging from data fetching and analysis to code quality, CI/CD, and business process automation, reflecting diverse organizational needs.
  • Nathaniel Whittemore argues that the sheer volume of AI-generated code will render traditional human code review unsustainable in 2026, making code quality and verification skills increasingly critical for enforcing standards and testing outputs.
  • Tariq identifies verification skills, which describe how to test or verify agent code, as having the highest ROI, suggesting that dedicating an engineer's week to perfect them is a valuable investment.
  • Anthropic's updated Skill Creator tool helps subject matter experts test, benchmark, and iteratively improve skills without coding, addressing issues like performance measurement, breakage with model updates, and skill triggering.
  • Ali Lemon noted that the Skill Creator's automatic description rewriting improved skill triggering five out of six times in Anthropic's internal tests, ensuring agents use skills effectively.
  • Anthropic categorizes skills into 'capability uplift' (doing new things) and 'encoded preference' (sequencing existing capabilities per workflow), noting that their testing needs and durability differ as models evolve.
  • Tariq's best practices for skill creation include avoiding the obvious, building a 'gotcha' section for common failure points, treating the entire file system as context engineering, and allowing Claude flexibility rather than railroading it.
  • For individual power users, skills act as 'reusable prompts with superpowers,' packaging code, templates, and data to ensure consistent, reliable agent performance for recurring tasks.
  • Notion AI has simplified the skills concept for mainstream users, allowing any page to be converted into a reusable skill, signaling a mental model shift from ad hoc prompting to reliable, repeatable AI capabilities across the stack.

RABBIT HOLE RECAP #405: STRETCH YOUR CHEEKS FOR THE BITCOIN BULLApr 17

  • At the OP_NEXT conference, institutional panelists from Coinbase and BlackRock expressed concern that investor uncertainty around Bitcoin's quantum resistance could limit capital inflows, a claim Marty finds ironic given Bitcoin's recent price surge.
  • Matt expresses a tinfoil-hat view that pressure for a quantum-related protocol change could be used to disenfranchise open-source developers and split the community, with institutions likely to push a fork that freezes legacy coins under the guise of an upgrade.
  • Odell highlights a new 'quantum-safe Bitcoin' proposal that uses existing consensus rules, requiring about $200 of GPU compute to create a safe address but making transactions non-standard. He likes that it provides an opt-in path without a soft fork.
  • Marty points out that Satoshi chose the libsecp256k1 cryptographic library because it lacked hard-coded constants that could hide a backdoor, arguing that blindly following NIST-approved standards for quantum resistance could introduce new vulnerabilities.
Also from this episode: (7)

Markets (3)

  • MicroStrategy raised roughly $2.1 billion via STRiPS this week, which Zach notes could be used to buy about 27,200 Bitcoin at current prices.
  • MicroStrategy's STRiPS currently trades at a slight discount, priced at $99.21 against its $100 par value, with a market cap of roughly $6.37 billion. Matt notes the product's dividend rate has climbed from its initial 9% to about 11.5%.
  • Matt argues the risk in STRiPS is layered and underappreciated, involving DeFi protocols, other public companies using it as a treasury asset, and the potential for a negative feedback loop if Bitcoin's price falls and MicroStrategy must sell shares to fund dividends.

Enterprise (1)

  • Michael Saylor proposed making STRiPS dividend payments semi-monthly instead of monthly, a change that would need shareholder approval. The hosts speculate this could smooth out the buying pressure around dividend dates.

Adoption (1)

  • Seth and Marty made a bet on whether MicroStrategy will hold over or under 1 million Bitcoin by June 15th, with Seth taking the under and Marty taking the over. MicroStrategy currently holds about 780,000 Bitcoin.

BTC Markets (1)

  • Arthur Hayes stated in an interview that over 90% of his net worth is in Bitcoin, leading the hosts to conclude many prominent 'shitcoiners' are actually Bitcoin maximalists using altcoins to accumulate more Bitcoin.

Custody (1)

  • Zach from BPI notes Tether's new self-custodial wallet is chain-agnostic and offers first-class Bitcoin and Lightning support, which he sees as a pragmatic step to onboard Tether users to Bitcoin.

The Resistance of a CowApr 17

Also from this episode: (12)

Biology (4)

  • Dairy farmer Gregers Christensen in Denmark reported his roughly 200 cows refused to drink from their water trough for months, instead drinking each other's urine and licking behinds to induce urination.
  • Journalist Klara Grunel found similar cases across Denmark and the US, like Minnesota dairy farmer Jill Nelson, who first noticed problems in 2008 with cows refusing to enter the milking parlor and drinking urine.
  • Veterinarian Don Sander argued cows drink urine primarily due to dietary mineral deficiencies like lack of potassium, especially in high-performance dairy breeds, and that lapping water is normal social or play behavior.
  • An unpublished 2016 Idaho study led by Richard Norell measured cow resistance at 200 ohms, but the data was lost and the 500-ohm standard remains supported by peer-reviewed research, leaving the debate unresolved.

Physics (3)

  • A Danish dowser named Gitte blamed the cows' behavior on stray electromagnetic energy from the neighboring Viking Link power station, which imports electricity from the UK, refusing to return to the farm.
  • Dairy electrician Larry Neubauer claims to have worked on nearly 4,000 to 5,000 stray voltage cases, describing it as a widespread problem that can destroy farms by causing cows to stop eating and drinking, leading to death.
  • The first stray voltage reports emerged in New Zealand in the 1960s, where barefoot farmers felt tingles, followed by North American cases in the 1970s and 1980s linked to new high-voltage power lines.

Energy (1)

  • In the 1970s, western Minnesota farmers protested new power lines over safety concerns, sometimes toppling towers, amid broader national unease about high-voltage infrastructure near farmland.

Science (2)

  • Professor Doug Reinemann's 1990s study for Wisconsin set the behavioral threshold for stray voltage at a level comparable to a 9-volt battery on a human tongue, finding less than 3% of over 9,000 investigated farms exceeded it.
  • The central scientific dispute is the cow's electrical resistance, set at 500 ohms by regulatory standards, but electricians and some researchers argue modern cows in wet, manure-rich barns have resistance as low as 200 ohms, making them more sensitive.

Health (2)

  • Professor Nigel Cook linked reported health issues to the 1990s shift from small tie-stall barns to large freestall operations, where herd management became more complex and individual cow monitoring declined.
  • Gregers Christensen ultimately sold his cows and switched to potato farming, while Jill Nelson, after intervention by electrician Larry Neubauer, saw her herd return to normal health.

Replit's CEO on Vibe Coding, Wealth Building, and What Most People Get Wrong About AIApr 15

  • Amjad Masad turned down a $1 billion acquisition offer for Replit when the company had six employees, believing he can build a trillion-dollar company instead.
  • Replit's revenue grew from $2.5 million to $250 million in just over a year. Its AI agent can now produce a working app in under an hour, shifting the platform from code-focused to fully automated.
  • Masad argues the primary bottleneck in the AI era is idea generation, not implementation. He cites an example where a finance guy using Replit built an app to automate investment banking tasks in one night and secured a $500k letter of intent the next day.
  • Masad says not having a coding background is becoming an advantage for founders because coders get lost in syntax, while product-focused people concentrate on marketing, UI, and solving the right problem.
  • He describes a concrete five-step process to build an app using AI: get a unique idea tied to a trend, break it down into a paragraph, focus on the core user journey for a 'five minute value' moment, use Replit to prompt-build the app, and iterate based on user feedback.
Also from this episode: (6)

Coding (1)

  • Masad's childhood in Jordan, where he built an internet cafe management system at age 13 and sold it for $500, inspired his mission to make coding accessible and a tool for wealth generation outside Silicon Valley.

AI & Tech (3)

  • He believes AI is not a job replacement but a tool for ambitious people to upgrade their workforce. The new high-value role is the 'generalist automator' who wields AI to find and fix company inefficiencies.
  • He rejects the AI doomer thesis, arguing mechanistic models cannot replicate human consciousness, inspiration, and the 'mystery of life' responsible for true paradigm-shifting discoveries.
  • Masad suggests improving communication with AI is not about special prompting but being a good general communicator, a skill you can develop through practices like improv, public speaking, and storytelling.

Business (1)

  • Masad views money as a fast-depreciating asset and advocates building wealth through equity ownership in businesses you start, join, or invest in, rather than holding cash or focusing on salary.

Psychology (1)

  • His ultimate advice for success is focused intention, perseverance, and the belief that 'no one is better than me,' which he credits for his initial achievements and ability to meet figures like Paul Graham, Sam Altman, and Tucker Carlson through a series of intentional connections.