Source Intelligence

DisclaimerUnofficial, and not affiliated with Anthropic. Nearly all of this is read straight out of what ships: npm bundles, captured prompts, published docs. Anthropic's own notes go in verbatim, marked as theirs. The rest is my reading, and every entry carries the strings behind it. If one looks wrong, vote it down and say why.

All releases Home olderv1.0.7 v1.0.9newer

Claude Code v1.0.8

16 entries read diff v1.0.7 → v1.0.8 Markdown
Find
Pick an entry · j / k steps through
9 entries

Changesopen

#

# Claude Code v1.0.8 Changelog

Related

Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.

MCP OAuth Authentication#

  • Added support for OAuth authentication with MCP (Model Context Protocol) servers
  • New mcp-auth command for authenticating with servers that require OAuth
  • Interactive authentication flow with browser-based login
  • Automatic token refresh when tokens expire
  • Secure credential storage in the configuration directory

Related

Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.

Enhanced Thinking Mode#

  • Expanded multi-language support for thinking mode triggers
  • Now supports thinking phrases in 8 languages:
  • English: "think", "think harder", "ultrathink", etc.
  • Japanese: "考えて", "熟考", "深く考えて"
  • Chinese: "思考", "深思", "仔细思考"
  • Spanish: "piensa", "piensa más", "piensa profundamente"
  • French: "réfléchis", "réfléchis plus", "pense"
  • German: "denk", "denk mehr", "denk gründlich"
  • Korean: "생각", "심사숙고", "깊이 생각"
  • Italian: "pensa", "rifletti", "pensa profondamente"
  • Three thinking levels with different token allocations:
  • HIGHEST: 31,999 tokens (phrases like "think harder", "ultrathink")
  • MIDDLE: 10,000 tokens (phrases like "think deeply", "megathink")
  • BASIC: 4,000 tokens (simple "think")

Related

Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.

Model Selection Improvements#

  • Better support for Claude Max users with automatic model selection
  • Simplified model detection for Pro vs Max subscriptions
  • Improved fallback model descriptions based on usage limits

Keyboard Shortcuts#

  • New Ctrl+E shortcut in transcript mode to toggle a feature (context-dependent)
  • Updated newline hints:
  • Apple Terminal with Option as Meta: "option + ⏎ for newline"
  • Other terminals: "shift + ⏎ for newline" or "\\⏎ for newline"

Related

Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.

Edit Tool Improvements#

  • Added replace_all parameter to the Edit tool
  • When replace_all is true, all occurrences of the old string are replaced
  • Default behavior remains single replacement for backwards compatibility

Example usage:

# Replace all occurrences of a variable name
claude "Rename all instances of 'oldVariable' to 'newVariable' in my code"
# Claude will use replace_all: true when appropriate

Signal Handling#

  • Improved process signal handling for SIGINT and SIGTERM
  • More graceful shutdown when interrupted

Related

Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.

Timeout Configuration#

  • Configurable MCP timeouts via environment variables:
  • MCP_TIMEOUT: General MCP operation timeout (default: 30 seconds)
  • MCP_TOOL_TIMEOUT: MCP tool execution timeout (default: 100 seconds)

Related

Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.

Debug Logging#

  • Enhanced MCP debug logging with --mcp-debug flag
  • Logs are written to structured JSON files for easier debugging
  • Each log entry includes timestamp, session ID, and working directory
1 entry

Bug Fixesopen

#

  • Fixed issues with file editing when strings end with newlines
  • Improved handling of partial JSON parsing for better error recovery
  • Better OAuth token refresh logic with proper error handling
  • Fixed race conditions in authentication flows

Related

Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.

1 entry

Internal Changesopen

#

  • Removed several unused functions and variables for cleaner codebase
  • Updated prompt caching logic to use function-based configuration
  • Improved configuration directory handling with hashing support
  • Better separation of Pro/Max user detection logic

Related

Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.

1 entry

Breaking Changesopen

#

None - this release maintains full backwards compatibility with v1.0.7.

Related

Other releases about the same thing. Found by shared names or similar wording; neither means one caused the other.

Verbatim
Official · Anthropic

Anthropic’s official release notes

Published verbatim by Anthropic for v1.0.8. Text is unmodified from the upstream changelog. Everything else on this page came out of the bundle instead, which is why the two lists don't match.

  • Fixed Vertex AI region fallback when using CLOUD_ML_REGION
  • Increased default otel interval from 1s -> 5s
  • Fixed edge cases where MCP_TIMEOUT and MCP_TOOL_TIMEOUT weren't being respected
  • Fixed a regression where search tools unnecessarily asked for permissions
  • Added support for triggering thinking non-English languages
  • Improved compacting UI
System prompt

The system prompt was not captured for this release, so this page cannot say whether it moved.