Claude Code

CodeDesktopWebGuidesMCPsSkillsToolsExtensions

A hidden Claude Code subagent watches another agent and reports drift. Observer agents: the enable flag, the digest, and the ObserverReport tool.

The Anthropic vs OpenAI race put frontier models in every ChatGPT and Claude tier. For moderate to heavy users, $100 on each lab beats $200 on one.

Claude Code runs four loop types: turn-based, goal-based, time-based, and proactive. What each means and how to set a real stop condition.

AdKit, Meta's official MCP, Pipeboard, Ryze, Composio compared: draft safety, pricing, OAuth, capabilities, and which to pick for what.

A second AI model that doesn't share Claude's blind spots catches more bugs than Claude reviewing itself. Here's how model stacking works.

What Claude Code /compact keeps and loses, straight from Anthropic's docs. The full survival table plus how to make rules and skills persist.

GLM 5.2 vs Opus 4.8 vs Sonnet 5: the only clean benchmark comparisons, why cross-vendor numbers mislead, pricing, open weights, and which to use.

GLM 5.2 is Z.ai's open-weights MIT model: 753B MoE, 1M context, $1.40/$4.40 API. Benchmarks, self-host reality, and how it stacks up vs Claude.

Sonnet 5 vs Opus 4.8 compared: Opus leads coding and reasoning by 0.5 to 6.6 points; Sonnet 5 ties knowledge work at ~40% less cost.

Claude Sonnet 5 scores 63.2% agentic coding at $2/$10 per million tokens (intro), closing the gap with Opus 4.8. Specs, benchmarks, pricing.

GPT-5.6 (Sol, Terra, Luna) is now generally available. Pricing, per-plan access, the launch-day reception, and how Sol compares to Claude.

Claude Code Artifacts turn session work into live, shareable pages at a versioned URL. Private by default, in beta for Team and Enterprise.

Claude Code sub-agents are now resumable via SendMessage and nest 5 levels deep. The persistent sub-agent pattern, field data, and the gotchas.

Claude Code subagents can now spawn their own subagents, capped at depth 5. How nesting works, the workflows it unlocks, and the token costs.

Is Claude Fable 5 free? It was, June 9-22, now usage credits at $10/$50 per million on the API. Pricing, the credits switch, and current access.

Claude Fable 5 benchmarks: state of the art on coding, finance, physics, and vision. Specs, the Mythos 5 split, safety, and release details.

Claude Code Dynamic Workflows vs Agent Teams: a side-by-side comparison and a clear rule for when to use each. Where ultracode fits.

Ultracode sends xhigh effort plus auto-triggered Dynamic Workflows. How it differs from xhigh, max, and ultrathink, plus the real token cost.

Why dynamic workflows exist, the three failure modes they fix, the six patterns, and how the harness runs. Beginner-friendly.

Claude Opus 4.8 API pricing ($5/$25 per 1M tokens), benchmarks, specs, and whether it is free. Grounded in Anthropic's official release.

Managing Claude Code teams needs one owner: the agent manager. The role, a DRI checklist, and a 90-day playbook for Claude Code at scale.

How to run Claude Code in monorepos, legacy codebases, and enterprise repos. 8 strategies for navigating millions of lines at scale.

Skills can be scoped to file paths. Workflows load only where they apply. Here is how, plus the rules vs workflows mental model.

One CLAUDE.md does not scale. Layer them per subdirectory. Claude walks the tree, loads the right rules, and your context stays lean.