{"version":"2.1.269","anchor":"mid-conversation-system-prompt-injection-now-restricted-to-f","canonical_anchor":"mid-conversation-system-prompt-injection-now-restricted-to-f","heading":"Mid-conversation system prompt injection now restricted to first-party provider","tier":"notice","area":"Elsewhere","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.269\/e\/mid-conversation-system-prompt-injection-now-restricted-to-f","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.269","markdown":"### Mid-conversation system prompt injection now restricted to first-party provider\n\nMid-conversation system prompt injection on tool changes now limited to first-party provider\n\n**What**\n\nThe internal check that decides whether to inject a mid-conversation system message when tools change now immediately returns false unless Claude Code is running against the first-party Anthropic provider. This means it now bails out for other providers, such as Bedrock or Vertex, before even checking the environment variable override that previously controlled this.\n\n**Why**\n\nThis stops Claude Code from injecting this system message on non-first-party providers, where it presumably wasn't intended or reliable to do so.\n\n- Area: Elsewhere\n- Tier: You'll notice\n- Useful: 2\/5\n- Signal: 2\/5"}