The best Claude Code skills: 24 picks from 9,578, plus the weird ones
The best Claude Code skills for guardrails, review, planning, and security, picked from 9,578 skills after reading each pick in full.
14 min · claude-codePutting AI agents to work in engineering teams: agentic coding, distributed systems, and the teams that ship them.
The best Claude Code skills for guardrails, review, planning, and security, picked from 9,578 skills after reading each pick in full.
14 min · claude-code
Claude Code headless brings your setup, background processes, and permission rules into a backend. The flags and live tests that caught the failures.
9 min · claude-code
When to use A2A: at a boundary where a standard contract pays off. Inside a tightly coupled fleet, it's usually overhead. A Docker test against MCP and NATS.
16 min · protocolsMost empty searches in my agent's memory were plumbing, not meaning. When not to use RAG: a week of logs, and bge-m3 against an LLM judge.
16 min · agentsThe best Claude Code skills for guardrails, review, planning, and security, picked from 9,578 skills after reading each pick in full.
Claude Code headless brings your setup, background processes, and permission rules into a backend. The flags and live tests that caught the failures.
When to use A2A: at a boundary where a standard contract pays off. Inside a tightly coupled fleet, it's usually overhead. A Docker test against MCP and NATS.
Most empty searches in my agent's memory were plumbing, not meaning. When not to use RAG: a week of logs, and bge-m3 against an LLM judge.
A Claude Code mod that keeps your agent's open questions above the prompt: about 250 lines, written by Claude itself, and how to build your own.
Claude Code hooks guide: all 33 lifecycle events, exit code 2 blocking, PreToolUse examples, the stop_hook_active trap, and the patterns worth shipping.
Claude Code SKILL.md frontmatter reference: all 20 fields, the context: fork and agent pair, body templating, and the bugs you'll hit in practice.
ForgeCode vs Claude Code: how ForgeCode's 81.8% Terminal-Bench score got rescored, where both agents stand on Terminal-Bench 4.0, and which CLI to pick now.
The Claude Code source leak, six months on: what the 2.1.88 source map exposed, which hidden features shipped, which died, and how the agent really works.
MCP vs A2A in 2026: what each AI agent protocol does, when you need A2A, and what happened to ACP, ANP, Agora, and the rest of the 2025 protocol rush.
Rust smart pointers cheat sheet: Box, Rc, Arc, RefCell, Cell, Cow, Mutex, and RwLock, with examples and a table of when to use each one.
Database replication methods compared: statement-based, write-ahead log (WAL), logical log, and trigger-based replication, with the pros and cons of each.
Kotter's 8-step change model in practice: an engineering leader's take on each step, from creating urgency to anchoring change in team culture.
Roll back PyCharm or any JetBrains IDE settings after a bad cloud sync: find the settingsSync Git history and restore your settings to a specific date.
Python threading, semaphores, and barriers explained with runnable code: limit concurrent workers, sync threads at a checkpoint, and avoid common pitfalls.
CAP theorem for developers: what a network partition forces you to choose, why CA isn't an option, how Spanner, DynamoDB, and Cassandra behave, and PACELC.
Solidity variable types explained: value vs reference types, storage, memory, and calldata, constant vs immutable, slot packing, and pragma versions.
Solidity gas optimization measured on solc 0.8.37: storage caching, calldata, custom errors, immutable, transient storage, optimizer settings, and the myths.
Pull over push in Solidity: why paying users in a loop breaks, and how to let them withdraw with call, checks-effects-interactions, and ReentrancyGuard.
Python deadlocks explained with examples: circular wait, self-deadlock, and starvation, plus 7 ways to avoid them, from lock ordering to timeouts.