v4.1.1

What's Changed

• 🧮 fix: Preserve Zod 4 Tool Schemas in Token Accounting by @lia-by-librechat[bot] in #642
• v4.1.1 by @lia-by-librechat[bot] in #643

Full Changelog: v4.1.0...v4.1.1

v4.1.0

What's Changed

• 🪨 fix: Strip Assistant Prefill for Claude 4.6+ on Bedrock by @berry-13 in #622
• 🪶 feat: Add Light Decisions and Rerank Subpath Exports by @danny-avila in #623
• 🎯 fix: Accept Two-Decimal Rounding and Quota Failures in System One Decisions by @danny-avila in #624
• 🔌 feat: Add Decision Credentials, a Breaker, and a Readiness Probe by @danny-avila in #625
• 🎚️ feat: Add Never-Throwing Rerank Outcomes With Per-Document System One, Cross-Encoder, BM25, and Fallback Chains by @danny-avila in #630
• 🧰 feat: Add Opt-In Tool Result Caps and Coding Tool Allowlist by @berry-13 in #612
• 🪢 fix: Preserve Decision and Rerank Lifecycle Invariants by @lia-by-librechat[bot] in #633
• v4.1.0 by @lia-by-librechat[bot] in #634

Full Changelog: v4.0.6...v4.1.0

v4.0.6

What's Changed

• 🔭 feat: Support Gemini 4 Argon Input and History Rules by @berry-13 in #620
• 🧯 fix: Strip Assistant Prefill for Claude 5 and Later Models by @berry-13 in #621

Full Changelog: v4.0.5...v4.0.6

v4.0.5

What's Changed

• 📭 fix: Reject Empty Provider Streams by @lia-by-librechat[bot] in #598
• 🧵 fix: Neutralize Harmony Control Tokens for OpenAI Models on Bedrock by @danny-avila in #611

Full Changelog: v4.0.4...v4.0.5

v4.0.4

What's Changed

• 🦉 feat: Add Read-Only GitHub Commit Comparison by @lia-by-librechat[bot] in #595

Full Changelog: v4.0.3...v4.0.4

v4.0.3

What's Changed

• 🎛️ feat: Add Typed Decision Models With HTTP and Structured Chat Adapters by @danny-avila in #561
• 🪜 feat: Add an Injectable Reranking Seam by @lia-by-librechat[bot] in #588
• 📡 fix: Expose Safe Subagent Startup Diagnostics by @lia-by-librechat[bot] in #592
• 🚦 feat: Gate Provider Text Before SDK Release by @lia-by-librechat[bot] in #593
• 🔌 fix: Verify Decision Output Bindings by Adapter by @lia-by-librechat[bot] in #590
• 🧭 feat: Let Hosts Declare Per-Call Subagent Arguments by @danny-avila in #596

Full Changelog: v4.0.1...v4.0.3

v4.0.1

What's Changed

• 🔦 fix: Log Why a Selected Subagent Could Not Start by @danny-avila in #583

Full Changelog: v4.0.0...v4.0.1

v4.0.0

What's Changed

• 🩹 fix: Don't Replay Reasoning to OpenAI Bedrock Models by @devanchohan in #581

New Contributors

• @devanchohan made their first contribution in #581

Full Changelog: v3.9.8...v4.0.0

v3.9.8

What's Changed

• ⏱️ feat: Expose Tool Preparation and Dispatch Timing by @lia-by-librechat[bot] in #577

Full Changelog: v3.9.7...v3.9.8

v3.9.7

What's Changed

• 🧪 test: Cover Subagent Depth-One Fan-Out And Continuations by @lia-by-librechat[bot] in #570
• ci: let the provider base URLs come from secrets by @danny-avila in #569
• 🛡️ fix: Let Background Subagents Run With HITL Enabled by @lia-by-librechat[bot] in #571
• 🏷️ fix: Scope Tool-Output Redaction to Tool Calls, Not Label Evidence by @danny-avila in #574
• 🩹 ci: Fix Invalid Workflow Blocking Package Publishing by @lia-by-librechat[bot] in #572

Full Changelog: v3.9.6...v3.9.7