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 olderv2.0.46 v2.0.49newer

Claude Code v2.0.47

5 entries read diff v2.0.46 → v2.0.47 Markdown

Version 2.0.47 introduces remote environment configuration for Teleport users, implements intelligent plan mode parallelism based on subscription tier, and adds automatic image compression to prevent API errors from oversized images.

Find
Pick an entry · j / k steps through
4 entries

Changesopen

Remote Environment Configuration#

What: New CLI command to configure default remote development environments for Teleport sessions

How to use:

# Launch interactive environment selector
claude remote-env

Details:

  • Interactive UI displays all available remote environments
  • Selected environment is saved to local settings as the default
  • Configuration persists across sessions
  • Environments can also be managed at https://claude.ai/code
  • Stores selection in settings.local.json under remote.defaultEnvironmentId
  • Evidence: EZ9() at line 489948, zZ9() at line 489984, command registration at line 490139

Automatic Image Compression#

What: Images in conversation context are now automatically compressed when they exceed size limits

How it works:

  • Automatically detects base64-encoded images larger than ~3.9MB
  • Compresses using JPEG format via Sharp library
  • Preserves original media type metadata
  • Prevents API errors from oversized content

Details:

  • Maximum uncompressed image size: 3,932,160 bytes (~3.9MB)
  • Compression threshold: 8,192 bytes
  • Only applies to images in conversation context, not uploaded files
  • Transparent to users - happens automatically during content processing
  • Evidence: egQ() at line 155776, called from content truncation at line 502920

Related

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

Plan Mode Exit State Tracking#

What: Internal state management to track when users have exited plan mode

Details:

  • Prevents re-prompting about plan mode after explicit exit
  • Improves workflow continuity in long sessions
  • State persists for the duration of the current session
  • Evidence: qV0() getter at line 2315, og() setter at line 2318

Plan Mode Parallelism Now Tier-Based#

What changed: Plan mode now adjusts parallel agent count based on subscription tier instead of using a fixed default

Previous behavior: All users got 4 parallel agents in plan mode (unless overridden by environment variable)

New behavior:

  • Free tier: 1 parallel agent (75% reduction)
  • Team/Enterprise tier: 3 parallel agents (25% reduction)
  • Max subscription with 20x model: 3 parallel agents
  • Environment variable override: Still available via CLAUDE_CODE_PLAN_V2_AGENT_COUNT (1-10)

Impact:

  • Free tier users will experience slower plan mode execution
  • Premium users retain better parallelism
  • Change encourages upgrade to paid tiers for better performance
  • Evidence: JQ2() at line 315736 (was v62() at line 316064 in v2.0.46)
2 entries

Bug Fixesopen

#

None identified in this release.

Related

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

OpenTelemetry Queue Configuration#

  • Added explicit max queue size constant (8,192) for telemetry batching
  • Improves observability infrastructure stability
  • No user-facing impact
Verbatim
Official · Anthropic

Anthropic’s official release notes

Published verbatim by Anthropic for v2.0.47. 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.

  • Improved error messages and validation for claude --teleport
  • Improved error handling in /usage
  • Fixed race condition with history entry not getting logged at exit
  • Fixed Vertex AI configuration not being applied from settings.json
System prompt

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