cc-practices-current

Refresh cached Claude Code feature and deprecation references from changelogs, releases, and npm metadata.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/leonardoacosta/central-claude --skill cc-practices-current
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cc-practices-current
Source: https://github.com/leonardoacosta/central-claude/tree/main/skills/cc-practices-current
Command: npx skills add https://github.com/leonardoacosta/central-claude --skill cc-practices-current

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires curl, jq, npm, sha256sum, timeout, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Keeps answers about Claude Code features, deprecations, and adoption guidance accurate by maintaining a continuously refreshed, diff-driven local cache so agents and audits do not rely on stale or noisy web searches.

Core Features & Use Cases

  • Deterministic refresh: A lean refresh.sh script fetches three upstream sources (docs changelog, GitHub releases, npm metadata), computes signatures, and signals when extraction is needed.
  • Structured references: When sources change, the skill rewrites concise, machine-friendly references (features, deprecations, adoption signals) that downstream agents use for audits and advice.
  • Safe economics: Diff-first design makes currency checks cheap and only invokes LLM extraction on actual upstream movement, reducing unnecessary work and false positives.
  • Integration: Powers /workflow:evolve and the cc-practices-analyst agent to provide verifiable, gradeable adoption checks and migration guidance.

Quick Start

Run the skill's refresh script and then ask the agent "what's new in Claude Code" to receive an authoritative, up-to-date summary.

Frequently Asked Questions about cc-practices-current

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I keep track of Claude Code features and deprecations without manual web searches?

Claude Code currency checks automate tracking by caching upstream sources like the docs changelog, GitHub releases, and npm metadata. A deterministic refresh script fetches these sources and only triggers LLM extraction when actual upstream changes are detected.

How does the diff-driven refresh mechanism work for Claude Code updates?

The diff-driven refresh mechanism computes signatures from fetched upstream sources and compares them against cached values. This diff-first design makes currency checks cheap by only invoking LLM extraction on actual movements, avoiding unnecessary work and false positives.

Do I need npm and curl installed to run automated Claude Code changelog audits?

Yes, running automated Claude Code changelog audits requires curl, jq, npm, sha256sum, and timeout installed in your environment. These dependencies support fetching upstream data, computing checksums, and processing JSON metadata.

What is the best way to check if a project setup is using deprecated Claude Code functionality?

Auditing project setups involves comparing local configurations against structured reference files of features and deprecations. The skill powers /workflow:evolve and the cc-practices-analyst agent to provide verifiable, gradeable adoption checks and migration guidance.

Can I integrate Claude Code release tracking into my existing automated workflow?

Yes, the skill exposes structured, machine-friendly reference files for downstream agents and automated checks. It powers /workflow:evolve and the cc-practices-analyst agent to provide verifiable adoption checks and migration guidance within existing pipelines.

When should I avoid using a local cache for Claude Code currency checks?

Relying on a local cache is not ideal if you need real-time, second-by-second updates rather than periodic deterministic refreshes. The system is designed for diff-driven efficiency, meaning it only updates references when upstream sources actually change.