Fix a specific problem

Targeted how-tos: prompt caching, bugfix loops, QA, tracing, setup drift.

40 cables · curated by @ragav · updated weekly
claude-code·no artifact

Anthropic's Responsible Scaling Policy (Sep 2023) — safety as operating procedure

*A five-part series tracing Anthropic's public thinking through Dario Amodei's writing and the company's model spec — one foundational document per entry, ea…

@frenxt · 6 mininstall →
claude-code·no artifact

Anthropic's "brilliant friend" spec — the product voice that defines Claude

*Part 2 of 5 — tracing Anthropic's public thinking with FRE|Nxt Labs production commentary.*

@frenxt · 5 mininstall →
claude-code·no artifact

Dario Amodei's Machines of Loving Grace (Oct 2024) — planning against the upside case

*Part 3 of 5 — tracing Anthropic's public thinking with FRE|Nxt Labs production commentary.*

@frenxt · 7 mininstall →
claude-code·no artifact

Dario Amodei's Urgency of Interpretability (April 2025) — the unsolved problem in production

*Part 4 of 5 — tracing Anthropic's public thinking with FRE|Nxt Labs production commentary.*

@frenxt · 7 mininstall →
claude-code·no artifact

Dario Amodei: "We are near the end of the exponential" (Feb 2026) — building for the endgame

*Final entry in the Amodei Playbook — from the safety floor of 2023 to the endgame framing of 2026.*

@frenxt · 7 mininstall →
claude-code·no artifact

Harrison Chase: "a single 800-line Python package" (Oct 2022) — the LangChain origin story

*A five-part series tracing Harrison Chase's public thinking on production AI agents — from LangChain as a weekend project to ambient agents — with FRE|Nxt L…

@frenxt · 5 mininstall →
claude-code·no artifact

Harrison Chase: LangGraph as the runtime (June 2024) — state machines for agents

*Part 2 of 5 — tracing Chase's production agent thinking with FRE|Nxt Labs commentary.*

@frenxt · 7 mininstall →
claude-code·no artifact

Harrison Chase: better models alone won't ship your agent (2025) — the production gap

*Part 3 of 5 — tracing Chase's production agent thinking with FRE|Nxt Labs commentary.*

@frenxt · 7 mininstall →
claude-code·no artifact

Harrison Chase: Deep Agents (July 2025) — batteries-included harness

*Part 4 of 5 — tracing Chase's production agent thinking with FRE|Nxt Labs commentary.*

@frenxt · 7 mininstall →
claude-code·no artifact

Harrison Chase: Ambient Agents (2025–26) — the UX after chat

*Final entry in the Chase Playbook — from an 800-line weekend project to the UX that replaces chat.*

@frenxt · 7 mininstall →
claude-code·no artifact

Boris Cherny: "code is not the bottleneck" (2024–26) — the origin story of Claude Code

*A four-part series breaking down the Claude Code creator's actual workflow — parallel agents, CLAUDE.md as postmortem, the "model wants tools" design philos…

@frenxt · 5 mininstall →
claude-code·no artifact

Boris Cherny: 5 parallel Claudes + 5-10 browser sessions (Jan 2026) — concurrency as the multiplier

*Part 2 of 4 — breaking down the Claude Code creator's actual workflow with FRE|Nxt Labs production commentary.*

@frenxt · 6 mininstall →
claude-code·no artifact

Boris Cherny: CLAUDE.md as a living postmortem (Jan 2026) — turning Claude into team memory

*Part 3 of 4 — breaking down the Claude Code creator's actual workflow with FRE|Nxt Labs production commentary.*

@frenxt · 6 mininstall →
claude-code·no artifact

Boris Cherny: "the model just wants to use tools" (2026) — the design philosophy behind Claude Code

*Final entry in the Cherny Playbook — from origin story to design philosophy.*

@frenxt · 7 mininstall →
cursor·no artifact

Set up Cursor Rules for your project

Cursor without rules is a capable assistant that doesn't know your project. Cursor with well-written rules is an assistant that already knows your stack, con…

@frenxt · 15 mininstall →
cursor·no artifact

Team Cursor Rules: shared conventions for multi-developer projects

The problem with AI-assisted teams is not that the tools are inconsistent — it's that each developer's tool has a different idea of what "correct" looks like…

@frenxt · 20 mininstall →
gemini-cli·no artifact

Your first GEMINI.md

Every session without a `GEMINI.md` starts from zero. You re-explain your stack. You re-explain your conventions. You catch mistakes that your context would…

@frenxt · 15 mininstall →
gemini-cli·no artifact

Install Gemini CLI and run your first prompt

The first time we ran Gemini CLI we expected it to behave like a browser-based chat. It doesn't. It's a terminal agent that reads your project, runs commands…

@frenxt · 10 mininstall →
gemini-cli·no artifact

Team GEMINI.md: shared AI conventions for multi-developer projects

When one engineer's Gemini CLI session produces different code shape than another's, the root cause is almost always context — not the tool. One person has a…

@frenxt · 20 mininstall →
claude-code·no artifact

Karpathy's Software 2.0 (2017) — the blog post that predicted the decade

*A six-part series tracing Andrej Karpathy's public thinking from 2017 to 2025 — one foundational post per entry, each with FRE|Nxt Labs live commentary on h…

@frenxt · 5 mininstall →
claude-code·no artifact

Karpathy's LLM OS (2023) — the architecture blueprint we still use today

*Part 2 of 6 — from Software 2.0 to the March of Nines, tracing Karpathy's public thinking with FRE|Nxt Labs production commentary.*

@frenxt · 7 mininstall →
claude-code·no artifact

Karpathy's Eureka Labs bet (2024) — what AI-native education reveals about his model of the future

*Part 3 of 6 — tracing Karpathy's public thinking with FRE|Nxt Labs production commentary.*

@frenxt · 5 mininstall →
claude-code·no artifact

Karpathy's vibe coding (Feb 2025) — and why "mostly works" is not a production SLA

*Part 4 of 6 — tracing Karpathy's public thinking with FRE|Nxt Labs production commentary.*

@frenxt · 5 mininstall →
claude-code·no artifact

Karpathy's Software 3.0 (June 2025) — "the hottest new programming language is English"

*Part 5 of 6 — tracing Karpathy's public thinking with FRE|Nxt Labs production commentary.*

@frenxt · 7 mininstall →
claude-code·no artifact

Karpathy's march of nines (Oct 2025) — why 90%→99.999% is the real AI problem

*Final entry in the Karpathy Playbook — from 2017's Software 2.0 to 2025's reality check.*

@frenxt · 7 mininstall →
qa-agent·no artifact

How QA Agent Works (Architecture)

Before contributing, understand what runs when you type `python cli.py run --suite smoke`.

@frenxt · 20 mininstall →
qa-agent·no artifact

Contributing to QA Agent

QA Agent is open source. Contributions that add integrations, improve the report format, or extend the agent's tool set are all welcome.

@frenxt · 25 mininstall →
qa-agent·no artifact

Writing Your First Test Case

A good test case reads like a conversation between a QA engineer and a developer — not like code.

@frenxt · 20 mininstall →
qa-agent·no artifact

Running Tests & Reading Reports

A green result is only useful if you trust it. A red result is only useful if you can debug it fast.

@frenxt · 20 mininstall →
qa-agent·no artifact

Setting Up QA Agent

From zero to running your first automated browser test in 15 minutes.

@frenxt · 15 mininstall →
qa-agent·no artifact

Organizing Suites & Personas

A flat folder of 30 test cases becomes unmanageable fast. Suites and personas are how you keep it structured.

@frenxt · 25 mininstall →
qa-agent·no artifact

What is QA Agent?

Code changes land fast. Your QA process shouldn't be the bottleneck.

@frenxt · 5 mininstall →
claude-code·skill

Publish to FRE|Nxt cables end-to-end

Publishing cables is not just writing markdown. The failures usually happen in validation, index generation, or submodule pointer updates.

@frenxt · 22 mininstall →
claude-code·skill

Replicate Ragav's stack (skills + plugins + scripts)

This cable installs the actual stack artifacts from Ragav's `~/.claude*` and `~/.codex` setup so anyone can replicate it quickly.

@frenxt · 20 mininstall →
claude-code·skill

Autonomous browser QA with browser-use

Manual QA catches issues, but it does not scale when UI changes land every day.

@frenxt · 28 mininstall →
claude-code·skill

Audit your E2E coverage

We had tests. We also had regressions. The problem was coverage shape, not raw test count.

@frenxt · 20 mininstall →
claude-code·skill

Benchmark a LangGraph agent

You cannot optimize what you do not measure. For agent systems, that means evaluating both quality and execution cost — and doing it per task class, not in a…

@frenxt · 35 mininstall →
claude-code·skill

Debug an agent from a LangSmith trace

When an agent fails, raw logs usually tell us *what* happened, not *why* it happened in that step sequence.

@frenxt · 25 mininstall →
claude-code·skill

Fix your prompt caching

Prompt caching is one of those optimizations we all think we enabled "well enough" until costs spike and response times drift.

@frenxt · 22 mininstall →
claude-code·skill

Reproduce -> fix -> verify

The fastest way to ship bugs is to skip one of those three steps.

@frenxt · 24 mininstall →