pricing
Pricing specified for Claude Opus 5.5: $4/$20 per million tokens with $0.20/Mtok cache reads
Users are informed of the cost structure for using Opus 5.5 model with prompt caching support.
Evidence
+$4/$20 per Mtok with $0.20/Mtok cache reads
Claude Code
Code assistantsclaude.com14 published changes
2 sources under watch
Change timeline
feature
Added AGENTS.md support as alternative to CLAUDE.md for project instructions
Users can now use AGENTS.md instead of CLAUDE.md for project instructions in Claude Code, configurable via /config panel (not yet available on Bedrock, Vertex, or Foundry)
Evidence
+Added AGENTS.md support: in a project with no CLAUDE.md, Claude Code reads AGENTS.md instead; change it under "Project instructions" in /config (not yet on Bedrock, Vertex or Foundry)
featurehigh
Added maxEffortLevel setting to cap effort level across all providers (Bedrock, Vertex, Foundry)
Users can now enforce a maximum effort level globally or per model, with the ability to select lower levels as needed. Affects organizations managing Claude Code settings.
Evidence
+Added `maxEffortLevel` setting (top-level or per model under `modelSettings`): caps the effort level on every provider, including Bedrock, Vertex and Foundry; users can still pick a lower level
api
Added CLAUDE_CODE_SUBAGENT_MODEL_FORCE environment variable for subagent model control
Users can now force all subagents to use a specific model via CLAUDE_CODE_SUBAGENT_MODEL_FORCE, overriding per-spawn and agent-definition model overrides.
Evidence
+Added <code>CLAUDE_CODE_SUBAGENT_MODEL_FORCE</code> to apply <code>CLAUDE_CODE_SUBAGENT_MODEL</code> (or the main model) to every subagent, ignoring per-spawn and agent-definition model overrides
featurehigh
Added Claude Fable 5.1 model with 1M context as new default Fable model
Users now have access to Claude Fable 5.1 (claude-fable-5-1) as the default Fable model with 1M context window, priced at $10/$50 per Mtok with $0.25/Mtok cache reads for improved performance.
Evidence
+Added Claude Fable 5.1 (<code>claude-fable-5-1</code>), now the default Fable model — 1M context, $10/$50 per Mtok with $0.25/Mtok cache reads
other
Fixed Bash commands failing on some Macs with task swap errors
Mac users running Bash commands will no longer encounter 'task output swap refused' errors when task directories are moved or linked.
Evidence
+Fixed Bash commands failing with "task output swap refused (tasks dir moved or linked)" on some Macs
other
Usage-limit message now includes reset time information for session and weekly limits
Users will see when their session or weekly spend limits reset when they reach their monthly limit.
Evidence
+The usage-limit message shown when your monthly spend limit is already used up now also says when your session or weekly limit resets
feature
Added ANTHROPIC_DEFAULT_MODEL environment variable for persistent model selection
Users can now set a default model for new sessions via environment variable, with /model picker still allowing overrides that persist across restarts
Evidence
+Added `ANTHROPIC_DEFAULT_MODEL` environment variable: sets the model new sessions start on, while a `/model` pick still overrides it and persists across restarts (unlike `ANTHROPIC_MODEL`)
other
Fixed MCP OAuth sign-in redirect URI mismatch issue for pre-registered OAuth clients
Developers using MCP with pre-registered OAuth clients like Slack can now successfully authenticate without redirect URI mismatches.
Evidence
+Fixed MCP OAuth sign-in failing with a redirect URI mismatch for servers that use a pre-registered OAuth client, such as Slack
positioning
Removed outdated note about auto mode sessions costing more for Pro, Max, and Team plans
Users on Pro, Max, and Team plans will no longer see a misleading note about auto mode pricing in first-use notices.
Evidence
+Removed the outdated note about auto mode sessions costing slightly more from the first-use notice for Pro, Max, and Team plans
otherhigh
Fixed Bash permission bypass vulnerability via crafted commands
Resolved a security vulnerability where crafted Bash commands could evade permission checks
Evidence
+Fixed a Bash permission bypass where a crafted command could hide parts of itself from permission checks
feature
Claude no longer auto-runs /verify and /code-review skills; must be explicitly invoked
Users must now explicitly invoke /verify and /code-review commands instead of Claude running them automatically, giving users more control over when these skills execute.
Evidence
+Claude no longer runs the `/verify` and `/code-review` skills on its own; invoke them with `/verify` or `/code-review` when you want them
api
Added OpenTelemetry log event attributes for message correlation and tool provenance
Developers can now use message.uuid, client_request_id, and tool_source attributes in OpenTelemetry logs to correlate messages and track tool origins.
Evidence
+Added `message.uuid`, `client_request_id`, and `tool_source` attributes to OpenTelemetry log events for message-level correlation and tool provenance
other
Multiple security fixes for permission checks and command validation
Fixed critical security issues including permission check bypasses in PowerShell, file-descriptor redirect handling, and auto-approval vulnerabilities in bash permission checks.
Evidence
+Fixed a permission-check bypass affecting commands run in Windows PowerShell 5.1 sessions; Fixed Bash permission checks to fail closed on file-descriptor redirect forms; Fixed Bash permission checks misjudging very long commands — commands over 10,000 characters now always prompt instead of running automatically