Sandbox
@marckrenn/claude-code-changelog

Claude Code prompt and flag history archive

This repo records Claude Code prompt artifacts, feature flags, and metadata across releases. It groups extracted prompt files in `system-prompts/`, adds summaries in `meta/`, and provides indexes for token count, init order, and last edit order.

885 stars71 forksUpdated 7d ago
Who it's for

People who want to study Claude Code release changes and prompt evolution.

What it delivers

You can compare Claude Code versions without digging through releases by hand.

What it does

Tracked release artifacts

Keeps legacy compatibility files like `cc-prompt.md` and `cc-flags.md` for each tag.

Extracted system prompts

Stores prompt artifacts in `system-prompts/*.md`, grouped by category.

Derived metadata

Summarizes flags, metadata, CLI surface, and prompt stats in `meta/`.

Prompt indexes

Provides index pages by token count, init time, and last edit time in `indices/`.

Release navigation

Uses annotated tags and GitHub releases to help you move through Claude Code history.

README

Claude Code Changelog

Status Uptime

[!NOTE] This archive takes some time and money (server, tokens) to run.

I’d appreciate any support – a free follow on X or chipping in a few bucks.

Follow on X GitHub Sponsors Buy Me a Coffee

Unofficial, community-maintained tracking of Claude Code prompt and feature-flag evolution across releases.

What's Tracked

  • cc-prompt.md – legacy compatibility file per tag
  • cc-flags.md – legacy compatibility file per tag
  • system-prompts/*.md – extracted prompt artifacts grouped by category (see indeces)
  • meta/flags.md, meta/metadata.md, meta/cli-surface.md, meta/prompt-stats.md – derived metadata
  • annotated tags (vX.Y.Z) and GitHub releases for historical navigation

How to Read This Repository

  1. Pick two tags and open compare view:
    • https://github.com/marckrenn/claude-code-changelog/compare/v2.1.44...v2.1.45#files_bucket
  2. Use system-prompts/ for per-file prompt artifacts.
  3. Use meta/ for structured summaries and aggregate metrics.
  4. Use Init / Last edit columns as lifecycle hints for each prompt artifact.
  5. For deep full-history research, point a coding agent at this repo directly. GitHub Copilot on github.com is currently limited for broad historical search/analysis.

Data Quality & Interpretation Notes

  • Token totals and per-file counts are estimates.
  • Tokenization can vary by model/tokenizer/runtime; small differences are expected.
  • Aggregate totals are useful directional signals.
  • Estimated LOC is derived from a prettified bundle line count and is only an approximation.
  • Prompt files can contain near-duplicates or intentionally duplicated variants.
  • File names can change across versions (renames/splits/merges), even when content lineage continues.
  • A single conceptual prompt can appear under multiple files in some releases.
  • Compare statistical deltas with raw diffs before drawing strong conclusions.

What Powers @ClaudeCodeLog

This repo is the backbone of the @ClaudeCodeLog X account, which automatically:

  1. Detects new Claude Code releases
  2. Extracts prompts and feature flags
  3. Generates structured diff summaries
  4. Publishes threaded updates with focused diff links and screenshots

Credits & Inspiration

System Prompts Index

Files in the repo

Repository payload8 top-level entries
  • .github
  • indices
  • meta
  • system-prompts
  • .gitattributes
  • cc-flags.md
  • cc-prompt.md
  • README.md

Discussion (0)

Ask about usage, or say what you built with it

Sign in to join the discussion.

No comments yet. Be the first to say what this is good for.

More other

HKUDS/
Vibe-Trading

"Vibe-Trading: Your Personal Trading Agent"

33k
tinyhumansai/
openhuman

OpenHuman is an open source personal AI for Mac, Windows and Linux — local-first memory, agent orchestration, and deep research.

40k

AIPOCH Open-Science is an open-source, local-first, model-agnostic AI research workbench for macOS, Windows, and Linux, with scientific agents, Python/R notebooks, data connectors, and reproducible provenance.

4k
eigent-ai/
eigent

Eigent: The Open Source Cowork Desktop - Local and Free Alternative to Claude Cowork and Codex

15k
op7418/
CodePilot

A multi-model AI agent desktop client — connect any AI provider, extend with MCP & skills, control from your phone. Built with Electron + Next.js.

6.5k