Bumps the npm-minor-patch group with 1 update in the /plugins/openclaw directory: [headroom-ai](https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript). Bumps the npm-minor-patch group with 3 updates in the /plugins/opencode directory: [headroom-ai](https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript), [effect](https://github.com/Effect-TS/effect/tree/HEAD/packages/effect) and [@opencode/plugin](https://github.com/anomalyco/opencode/tree/HEAD/packages/plugin). Bumps the npm-minor-patch group with 7 updates in the /sdk/typescript directory: | Package | From | To | | --- | --- | --- | | [@ai-sdk/anthropic](https://github.com/vercel/ai/tree/HEAD/packages/anthropic) | `4.0.94` | `4.0.99` | | [@ai-sdk/openai](https://github.com/vercel/ai/tree/HEAD/packages/openai) | `4.0.106` | `4.0.111` | | [@ai-sdk/provider](https://github.com/vercel/ai/tree/HEAD/packages/provider) | `4.0.35` | `4.0.37` | | [@anthropic-ai/sdk](https://github.com/anthropics/anthropic-sdk-typescript) | `1.3.0` | `1.6.0` | | [ai](https://github.com/vercel/ai/tree/HEAD/packages/ai) | `7.0.159` | `7.0.166` | | [dotenv](https://github.com/motdotla/dotenv) | `18.0.9` | `18.0.10` | | [openai](https://github.com/openai/openai-node) | `7.47.0` | `7.53.0` | Updates `headroom-ai` from 0.42.0 to 0.43.1 <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/headroomlabs-ai/headroom/releases">headroom-ai's releases</a>.</em></p> <blockquote> <h2>Release v0.43.1</h2> <h2><a href="https://github.com/headroomlabs-ai/headroom/compare/v0.43.0...v0.43.1">0.43.1</a> (2026-11-05)</h2> <h3>Bug Fixes</h3> <ul> <li><strong>ccr:</strong> keep the in-memory order queue from evicting re-stored entries (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4545">#4545</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/c6f5bf9549e9b357e9c2369d4e04bc437f5e180d">c6f5bf9</a>)</li> <li><strong>memory:</strong> read the relation key the memory_save schema advertises (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4647">#4647</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/7592e681459f14a2041eb5f524a658cb6ad1dd92">7592e68</a>)</li> <li><strong>memory:</strong> replace the HNSW vector when a memory is re-indexed (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4649">#4649</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/ee638f9532b3788bacd13ed8d26dd0faf0740baf">ee638f9</a>)</li> <li><strong>openclaw:</strong> accept 12-character hashes in headroom_retrieve (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4640">#4640</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/8d3e9e5ca7cd645eb73084b848edbea5fc796e1a">8d3e9e5</a>)</li> <li><strong>ort:</strong> avoid implicit runtime discovery from the workspace (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4570">#4570</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/69fbd2390d9c8b0fa1f59fb8314bc9b8f01b1a55">69fbd23</a>)</li> <li><strong>proxy:</strong> bound the model-keyed tokenizer, beacon and litellm caches (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4497">#4497</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e51b8065a4a2aa3a3c7f266c1cb8d8bde6b568d0">e51b806</a>)</li> <li><strong>proxy:</strong> use HTTP/1.1 client for ChatGPT streaming and model metadata (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4006">#4006</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e1becdf7329bf4ca03874b9f512ac19837ca29c2">e1becdf</a>)</li> <li><strong>release:</strong> normalize damaged release PR envelopes before detection (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4664">#4664</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/917edc62230c3c4a8a415e2053da68dab1f003dd">917edc6</a>)</li> <li><strong>router:</strong> keep tagged blocks out of the lossy size-gate fallback (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4543">#4543</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/f333edf20f0d7acf5ea67e040c1f58539a0edd4d">f333edf</a>)</li> <li><strong>router:</strong> log the ONNX fallbacks at info instead of warning (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4535">#4535</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/db30d82dfffdb3bbe4d60a0e0bfe88640e58470f">db30d82</a>)</li> </ul> <h3>Dependencies</h3> <ul> <li>bump the npm-minor-patch group across 3 directories with 6 updates (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4662">#4662</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/8f104f8e298c85ff0005f2fd4848eea95f32b7c7">8f104f8</a>)</li> </ul> <!-- raw HTML omitted --> <hr /> <p><strong>Want more savings?</strong> Headroom Enterprise saves more for teams and enterprises. Talk to us at <a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/blob/HEAD/mailto:hello@headroomlabs.ai">hello@headroomlabs.ai</a> or <a href="https://headroomlabs.ai">headroomlabs.ai</a>.</p> <h2>Release v0.43.0</h2> <h2><a href="https://github.com/headroomlabs-ai/headroom/compare/v0.42.0...v0.43.0">0.43.0</a> (2026-11-05)</h2> <h3>⚠ BREAKING CHANGES</h3> <ul> <li><strong>mcp:</strong> port MCP servers to SDK 2.x (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4603">#4603</a>)</li> </ul> <h3>Features</h3> <ul> <li><strong>ccr:</strong> answer headroom_retrieve on streaming chat completions (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4631">#4631</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/31dcaf41f7fafaebec0838a05ff328da456faef0">31dcaf4</a>)</li> <li><strong>mcp:</strong> port MCP servers to SDK 2.x (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4603">#4603</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/ee2252a3155c87e948c8893b79af5824eb30e64c">ee2252a</a>)</li> <li><strong>plugins:</strong> headroom skill for Claude Code, Codex, Copilot and skills.sh (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4533">#4533</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/66e14676827366c2bf8d15e252f8766cb27b1a98">66e1467</a>)</li> <li><strong>savings:</strong> show the proxy's model-routing savings in headroom savings (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4304">#4304</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/aaab338b64cb55929de904411db65db126e22ec0">aaab338</a>)</li> </ul> <h3>Bug Fixes</h3> <ul> <li>[BUG] A timeout in one image compression call resets the shared worker pool and affects concurrent requests (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4528">#4528</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e9bcfe73fadf69f0b27b7bc338ea7d2380931054">e9bcfe7</a>)</li> <li>[BUG] DirectMem0Adapter.search_memories calls removed mem0 search API (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4518">#4518</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/d9cee452681c79d6f13919e9e651e88a774c0874">d9cee45</a>)</li> <li>[BUG] Saved request metrics lose their cache optimization fields (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4584">#4584</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/b4c9bdf100d97fb038beba95de6d4ad6a8786090">b4c9bdf</a>)</li> <li>[BUG] Volatile SQLite metrics stores lose their schema before the first save (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4633">#4633</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/d8ff0fe9d2cb2477f6662a90f8fcd7e030a52666">d8ff0fe</a>)</li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Changelog</summary> <p><em>Sourced from <a href="https://github.com/headroomlabs-ai/headroom/blob/main/CHANGELOG.md">headroom-ai's changelog</a>.</em></p> <blockquote> <h2><a href="https://github.com/headroomlabs-ai/headroom/compare/v0.43.0...v0.43.1">0.43.1</a> (2026-11-05)</h2> <h3>Bug Fixes</h3> <ul> <li><strong>ccr:</strong> keep the in-memory order queue from evicting re-stored entries (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4545">#4545</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/c6f5bf9549e9b357e9c2369d4e04bc437f5e180d">c6f5bf9</a>)</li> <li><strong>memory:</strong> read the relation key the memory_save schema advertises (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4647">#4647</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/7592e681459f14a2041eb5f524a658cb6ad1dd92">7592e68</a>)</li> <li><strong>memory:</strong> replace the HNSW vector when a memory is re-indexed (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4649">#4649</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/ee638f9532b3788bacd13ed8d26dd0faf0740baf">ee638f9</a>)</li> <li><strong>openclaw:</strong> accept 12-character hashes in headroom_retrieve (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4640">#4640</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/8d3e9e5ca7cd645eb73084b848edbea5fc796e1a">8d3e9e5</a>)</li> <li><strong>ort:</strong> avoid implicit runtime discovery from the workspace (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4570">#4570</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/69fbd2390d9c8b0fa1f59fb8314bc9b8f01b1a55">69fbd23</a>)</li> <li><strong>proxy:</strong> bound the model-keyed tokenizer, beacon and litellm caches (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4497">#4497</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e51b8065a4a2aa3a3c7f266c1cb8d8bde6b568d0">e51b806</a>)</li> <li><strong>proxy:</strong> use HTTP/1.1 client for ChatGPT streaming and model metadata (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4006">#4006</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e1becdf7329bf4ca03874b9f512ac19837ca29c2">e1becdf</a>)</li> <li><strong>release:</strong> normalize damaged release PR envelopes before detection (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4664">#4664</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/917edc62230c3c4a8a415e2053da68dab1f003dd">917edc6</a>)</li> <li><strong>router:</strong> keep tagged blocks out of the lossy size-gate fallback (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4543">#4543</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/f333edf20f0d7acf5ea67e040c1f58539a0edd4d">f333edf</a>)</li> <li><strong>router:</strong> log the ONNX fallbacks at info instead of warning (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4535">#4535</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/db30d82dfffdb3bbe4d60a0e0bfe88640e58470f">db30d82</a>)</li> </ul> <h3>Dependencies</h3> <ul> <li>bump the npm-minor-patch group across 3 directories with 6 updates (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4662">#4662</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/8f104f8e298c85ff0005f2fd4848eea95f32b7c7">8f104f8</a>)</li> </ul> <h2><a href="https://github.com/headroomlabs-ai/headroom/compare/v0.42.0...v0.43.0">0.43.0</a> (2026-11-05)</h2> <h3>⚠ BREAKING CHANGES</h3> <ul> <li><strong>mcp:</strong> port MCP servers to SDK 2.x (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4603">#4603</a>)</li> </ul> <h3>Features</h3> <ul> <li><strong>ccr:</strong> answer headroom_retrieve on streaming chat completions (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4631">#4631</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/31dcaf41f7fafaebec0838a05ff328da456faef0">31dcaf4</a>)</li> <li><strong>mcp:</strong> port MCP servers to SDK 2.x (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4603">#4603</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/ee2252a3155c87e948c8893b79af5824eb30e64c">ee2252a</a>)</li> <li><strong>plugins:</strong> headroom skill for Claude Code, Codex, Copilot and skills.sh (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4533">#4533</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/66e14676827366c2bf8d15e252f8766cb27b1a98">66e1467</a>)</li> <li><strong>savings:</strong> show the proxy's model-routing savings in headroom savings (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4304">#4304</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/aaab338b64cb55929de904411db65db126e22ec0">aaab338</a>)</li> </ul> <h3>Bug Fixes</h3> <ul> <li>[BUG] A timeout in one image compression call resets the shared worker pool and affects concurrent requests (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4528">#4528</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e9bcfe73fadf69f0b27b7bc338ea7d2380931054">e9bcfe7</a>)</li> <li>[BUG] DirectMem0Adapter.search_memories calls removed mem0 search API (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4518">#4518</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/d9cee452681c79d6f13919e9e651e88a774c0874">d9cee45</a>)</li> <li>[BUG] Saved request metrics lose their cache optimization fields (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4584">#4584</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/b4c9bdf100d97fb038beba95de6d4ad6a8786090">b4c9bdf</a>)</li> <li>[BUG] Volatile SQLite metrics stores lose their schema before the first save (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4633">#4633</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/d8ff0fe9d2cb2477f6662a90f8fcd7e030a52666">d8ff0fe</a>)</li> <li>Add info about apply to README.md (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/1592">#1592</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/4d34680c34ad379d693820869e77d4a2a6be2cdf">4d34680</a>)</li> <li><strong>backends/litellm:</strong> keep cache_control on text blocks for Bedrock prompt caching (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4565">#4565</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/23a5bb1c81dc943398f9271f068a02487207f2e7">23a5bb1</a>)</li> <li><strong>backends:</strong> forward the standard service_tier param (litellm + anyllm) (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/2468">#2468</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/62cc38a6b09dbcf723681a0ff56acd02131ad1f1">62cc38a</a>)</li> <li><strong>cache:</strong> build semantic entry before evicting so a failed embedding keeps live entries (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4581">#4581</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/0d4c450d774691568b850f0d22ce17e087405281">0d4c450</a>)</li> <li><strong>cache:</strong> count a tool_result with no text as replayable in the frozen prefix (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4523">#4523</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/fba09f674c8a9c91b39a563a70ab7727b44d1dbc">fba09f6</a>)</li> <li><strong>cache:</strong> fail closed on SQLite corruption instead of deleting the store online (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4361">#4361</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/98018333f997783487af15efe4d9e631ecd85939">9801833</a>)</li> <li><strong>cache:</strong> invalidate configured registry instances on replacement (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4363">#4363</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/1bcdca2d9dd279f49aa4276d523d93c2cfeb5a51">1bcdca2</a>)</li> <li><strong>cache:</strong> namespace SemanticCacheLayer by provider, model and metadata (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4473">#4473</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/99cdaaf1c7de2bd81a8e64f198462fd1ceec3a17">99cdaaf</a>)</li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/2d3ee51846435ef5b320eea4806af14e602b0d8c"><code>2d3ee51</code></a> chore: release 0.43.1 (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4661">#4661</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/8f104f8e298c85ff0005f2fd4848eea95f32b7c7"><code>8f104f8</code></a> deps: bump the npm-minor-patch group across 3 directories with 6 updates (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4662">#4662</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/81fa4bdda613f6a37f6f430eda9f728e1458f3ce"><code>81fa4bd</code></a> chore: release 0.43.0 (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4643">#4643</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/a26faaeee34ecfde57608aea4a1ec7ea2b33272f"><code>a26faae</code></a> fix(sdk): carry Vercel AI SDK reasoning parts and provider options through co...</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/e666fd8e35dbe7c0433739297873b6a0a51c26cf"><code>e666fd8</code></a> deps: Bump tsup from 8.5.1 to 9.0.0 in /sdk/typescript (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4386">#4386</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/74277d47ecfbca349ddb99f81094c4dd30539d43"><code>74277d4</code></a> deps: Bump <code>@anthropic-ai/sdk</code> from 0.133.0 to 1.3.0 in /sdk/typescript (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4554">#4554</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/5e4827de12c9fdf33d7d36b1939eee73a18fe6c5"><code>5e4827d</code></a> deps: Bump the npm-minor-patch group across 1 directory with 6 updates (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4621">#4621</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/cd5e2b9e73eb8bd783ab872bdda71263b941e2d3"><code>cd5e2b9</code></a> fix(sdk): send Anthropic messages to /v1/compress natively so thinking blocks...</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/6c3d7475077b2e09e8cff5ba2352bda4064d86d3"><code>6c3d747</code></a> chore(sdk-typescript): npm metadata and the proxy requirement up front (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4536">#4536</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/f36772805178e3fed1a95313e8211c57a291b953"><code>f367728</code></a> deps: Bump the npm-minor-patch group across 3 directories with 9 updates (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4495">#4495</a>)</li> <li>See full diff in <a href="https://github.com/headroomlabs-ai/headroom/commits/v0.43.1/sdk/typescript">compare view</a></li> </ul> </details> <br /> Updates `headroom-ai` from 0.42.0 to 0.43.1 <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/headroomlabs-ai/headroom/releases">headroom-ai's releases</a>.</em></p> <blockquote> <h2>Release v0.43.1</h2> <h2><a href="https://github.com/headroomlabs-ai/headroom/compare/v0.43.0...v0.43.1">0.43.1</a> (2026-11-05)</h2> <h3>Bug Fixes</h3> <ul> <li><strong>ccr:</strong> keep the in-memory order queue from evicting re-stored entries (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4545">#4545</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/c6f5bf9549e9b357e9c2369d4e04bc437f5e180d">c6f5bf9</a>)</li> <li><strong>memory:</strong> read the relation key the memory_save schema advertises (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4647">#4647</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/7592e681459f14a2041eb5f524a658cb6ad1dd92">7592e68</a>)</li> <li><strong>memory:</strong> replace the HNSW vector when a memory is re-indexed (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4649">#4649</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/ee638f9532b3788bacd13ed8d26dd0faf0740baf">ee638f9</a>)</li> <li><strong>openclaw:</strong> accept 12-character hashes in headroom_retrieve (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4640">#4640</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/8d3e9e5ca7cd645eb73084b848edbea5fc796e1a">8d3e9e5</a>)</li> <li><strong>ort:</strong> avoid implicit runtime discovery from the workspace (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4570">#4570</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/69fbd2390d9c8b0fa1f59fb8314bc9b8f01b1a55">69fbd23</a>)</li> <li><strong>proxy:</strong> bound the model-keyed tokenizer, beacon and litellm caches (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4497">#4497</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e51b8065a4a2aa3a3c7f266c1cb8d8bde6b568d0">e51b806</a>)</li> <li><strong>proxy:</strong> use HTTP/1.1 client for ChatGPT streaming and model metadata (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4006">#4006</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e1becdf7329bf4ca03874b9f512ac19837ca29c2">e1becdf</a>)</li> <li><strong>release:</strong> normalize damaged release PR envelopes before detection (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4664">#4664</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/917edc62230c3c4a8a415e2053da68dab1f003dd">917edc6</a>)</li> <li><strong>router:</strong> keep tagged blocks out of the lossy size-gate fallback (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4543">#4543</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/f333edf20f0d7acf5ea67e040c1f58539a0edd4d">f333edf</a>)</li> <li><strong>router:</strong> log the ONNX fallbacks at info instead of warning (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4535">#4535</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/db30d82dfffdb3bbe4d60a0e0bfe88640e58470f">db30d82</a>)</li> </ul> <h3>Dependencies</h3> <ul> <li>bump the npm-minor-patch group across 3 directories with 6 updates (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4662">#4662</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/8f104f8e298c85ff0005f2fd4848eea95f32b7c7">8f104f8</a>)</li> </ul> <!-- raw HTML omitted --> <hr /> <p><strong>Want more savings?</strong> Headroom Enterprise saves more for teams and enterprises. Talk to us at <a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/blob/HEAD/mailto:hello@headroomlabs.ai">hello@headroomlabs.ai</a> or <a href="https://headroomlabs.ai">headroomlabs.ai</a>.</p> <h2>Release v0.43.0</h2> <h2><a href="https://github.com/headroomlabs-ai/headroom/compare/v0.42.0...v0.43.0">0.43.0</a> (2026-11-05)</h2> <h3>⚠ BREAKING CHANGES</h3> <ul> <li><strong>mcp:</strong> port MCP servers to SDK 2.x (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4603">#4603</a>)</li> </ul> <h3>Features</h3> <ul> <li><strong>ccr:</strong> answer headroom_retrieve on streaming chat completions (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4631">#4631</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/31dcaf41f7fafaebec0838a05ff328da456faef0">31dcaf4</a>)</li> <li><strong>mcp:</strong> port MCP servers to SDK 2.x (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4603">#4603</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/ee2252a3155c87e948c8893b79af5824eb30e64c">ee2252a</a>)</li> <li><strong>plugins:</strong> headroom skill for Claude Code, Codex, Copilot and skills.sh (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4533">#4533</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/66e14676827366c2bf8d15e252f8766cb27b1a98">66e1467</a>)</li> <li><strong>savings:</strong> show the proxy's model-routing savings in headroom savings (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4304">#4304</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/aaab338b64cb55929de904411db65db126e22ec0">aaab338</a>)</li> </ul> <h3>Bug Fixes</h3> <ul> <li>[BUG] A timeout in one image compression call resets the shared worker pool and affects concurrent requests (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4528">#4528</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e9bcfe73fadf69f0b27b7bc338ea7d2380931054">e9bcfe7</a>)</li> <li>[BUG] DirectMem0Adapter.search_memories calls removed mem0 search API (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4518">#4518</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/d9cee452681c79d6f13919e9e651e88a774c0874">d9cee45</a>)</li> <li>[BUG] Saved request metrics lose their cache optimization fields (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4584">#4584</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/b4c9bdf100d97fb038beba95de6d4ad6a8786090">b4c9bdf</a>)</li> <li>[BUG] Volatile SQLite metrics stores lose their schema before the first save (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4633">#4633</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/d8ff0fe9d2cb2477f6662a90f8fcd7e030a52666">d8ff0fe</a>)</li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Changelog</summary> <p><em>Sourced from <a href="https://github.com/headroomlabs-ai/headroom/blob/main/CHANGELOG.md">headroom-ai's changelog</a>.</em></p> <blockquote> <h2><a href="https://github.com/headroomlabs-ai/headroom/compare/v0.43.0...v0.43.1">0.43.1</a> (2026-11-05)</h2> <h3>Bug Fixes</h3> <ul> <li><strong>ccr:</strong> keep the in-memory order queue from evicting re-stored entries (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4545">#4545</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/c6f5bf9549e9b357e9c2369d4e04bc437f5e180d">c6f5bf9</a>)</li> <li><strong>memory:</strong> read the relation key the memory_save schema advertises (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4647">#4647</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/7592e681459f14a2041eb5f524a658cb6ad1dd92">7592e68</a>)</li> <li><strong>memory:</strong> replace the HNSW vector when a memory is re-indexed (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4649">#4649</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/ee638f9532b3788bacd13ed8d26dd0faf0740baf">ee638f9</a>)</li> <li><strong>openclaw:</strong> accept 12-character hashes in headroom_retrieve (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4640">#4640</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/8d3e9e5ca7cd645eb73084b848edbea5fc796e1a">8d3e9e5</a>)</li> <li><strong>ort:</strong> avoid implicit runtime discovery from the workspace (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4570">#4570</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/69fbd2390d9c8b0fa1f59fb8314bc9b8f01b1a55">69fbd23</a>)</li> <li><strong>proxy:</strong> bound the model-keyed tokenizer, beacon and litellm caches (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4497">#4497</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e51b8065a4a2aa3a3c7f266c1cb8d8bde6b568d0">e51b806</a>)</li> <li><strong>proxy:</strong> use HTTP/1.1 client for ChatGPT streaming and model metadata (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4006">#4006</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e1becdf7329bf4ca03874b9f512ac19837ca29c2">e1becdf</a>)</li> <li><strong>release:</strong> normalize damaged release PR envelopes before detection (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4664">#4664</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/917edc62230c3c4a8a415e2053da68dab1f003dd">917edc6</a>)</li> <li><strong>router:</strong> keep tagged blocks out of the lossy size-gate fallback (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4543">#4543</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/f333edf20f0d7acf5ea67e040c1f58539a0edd4d">f333edf</a>)</li> <li><strong>router:</strong> log the ONNX fallbacks at info instead of warning (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4535">#4535</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/db30d82dfffdb3bbe4d60a0e0bfe88640e58470f">db30d82</a>)</li> </ul> <h3>Dependencies</h3> <ul> <li>bump the npm-minor-patch group across 3 directories with 6 updates (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4662">#4662</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/8f104f8e298c85ff0005f2fd4848eea95f32b7c7">8f104f8</a>)</li> </ul> <h2><a href="https://github.com/headroomlabs-ai/headroom/compare/v0.42.0...v0.43.0">0.43.0</a> (2026-11-05)</h2> <h3>⚠ BREAKING CHANGES</h3> <ul> <li><strong>mcp:</strong> port MCP servers to SDK 2.x (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4603">#4603</a>)</li> </ul> <h3>Features</h3> <ul> <li><strong>ccr:</strong> answer headroom_retrieve on streaming chat completions (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4631">#4631</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/31dcaf41f7fafaebec0838a05ff328da456faef0">31dcaf4</a>)</li> <li><strong>mcp:</strong> port MCP servers to SDK 2.x (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4603">#4603</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/ee2252a3155c87e948c8893b79af5824eb30e64c">ee2252a</a>)</li> <li><strong>plugins:</strong> headroom skill for Claude Code, Codex, Copilot and skills.sh (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4533">#4533</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/66e14676827366c2bf8d15e252f8766cb27b1a98">66e1467</a>)</li> <li><strong>savings:</strong> show the proxy's model-routing savings in headroom savings (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4304">#4304</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/aaab338b64cb55929de904411db65db126e22ec0">aaab338</a>)</li> </ul> <h3>Bug Fixes</h3> <ul> <li>[BUG] A timeout in one image compression call resets the shared worker pool and affects concurrent requests (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4528">#4528</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/e9bcfe73fadf69f0b27b7bc338ea7d2380931054">e9bcfe7</a>)</li> <li>[BUG] DirectMem0Adapter.search_memories calls removed mem0 search API (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4518">#4518</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/d9cee452681c79d6f13919e9e651e88a774c0874">d9cee45</a>)</li> <li>[BUG] Saved request metrics lose their cache optimization fields (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4584">#4584</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/b4c9bdf100d97fb038beba95de6d4ad6a8786090">b4c9bdf</a>)</li> <li>[BUG] Volatile SQLite metrics stores lose their schema before the first save (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4633">#4633</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/d8ff0fe9d2cb2477f6662a90f8fcd7e030a52666">d8ff0fe</a>)</li> <li>Add info about apply to README.md (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/1592">#1592</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/4d34680c34ad379d693820869e77d4a2a6be2cdf">4d34680</a>)</li> <li><strong>backends/litellm:</strong> keep cache_control on text blocks for Bedrock prompt caching (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4565">#4565</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/23a5bb1c81dc943398f9271f068a02487207f2e7">23a5bb1</a>)</li> <li><strong>backends:</strong> forward the standard service_tier param (litellm + anyllm) (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/2468">#2468</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/62cc38a6b09dbcf723681a0ff56acd02131ad1f1">62cc38a</a>)</li> <li><strong>cache:</strong> build semantic entry before evicting so a failed embedding keeps live entries (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4581">#4581</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/0d4c450d774691568b850f0d22ce17e087405281">0d4c450</a>)</li> <li><strong>cache:</strong> count a tool_result with no text as replayable in the frozen prefix (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4523">#4523</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/fba09f674c8a9c91b39a563a70ab7727b44d1dbc">fba09f6</a>)</li> <li><strong>cache:</strong> fail closed on SQLite corruption instead of deleting the store online (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4361">#4361</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/98018333f997783487af15efe4d9e631ecd85939">9801833</a>)</li> <li><strong>cache:</strong> invalidate configured registry instances on replacement (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4363">#4363</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/1bcdca2d9dd279f49aa4276d523d93c2cfeb5a51">1bcdca2</a>)</li> <li><strong>cache:</strong> namespace SemanticCacheLayer by provider, model and metadata (<a href="https://redirect.github.com/headroomlabs-ai/headroom/issues/4473">#4473</a>) (<a href="https://github.com/headroomlabs-ai/headroom/commit/99cdaaf1c7de2bd81a8e64f198462fd1ceec3a17">99cdaaf</a>)</li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/2d3ee51846435ef5b320eea4806af14e602b0d8c"><code>2d3ee51</code></a> chore: release 0.43.1 (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4661">#4661</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/8f104f8e298c85ff0005f2fd4848eea95f32b7c7"><code>8f104f8</code></a> deps: bump the npm-minor-patch group across 3 directories with 6 updates (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4662">#4662</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/81fa4bdda613f6a37f6f430eda9f728e1458f3ce"><code>81fa4bd</code></a> chore: release 0.43.0 (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4643">#4643</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/a26faaeee34ecfde57608aea4a1ec7ea2b33272f"><code>a26faae</code></a> fix(sdk): carry Vercel AI SDK reasoning parts and provider options through co...</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/e666fd8e35dbe7c0433739297873b6a0a51c26cf"><code>e666fd8</code></a> deps: Bump tsup from 8.5.1 to 9.0.0 in /sdk/typescript (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4386">#4386</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/74277d47ecfbca349ddb99f81094c4dd30539d43"><code>74277d4</code></a> deps: Bump <code>@anthropic-ai/sdk</code> from 0.133.0 to 1.3.0 in /sdk/typescript (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4554">#4554</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/5e4827de12c9fdf33d7d36b1939eee73a18fe6c5"><code>5e4827d</code></a> deps: Bump the npm-minor-patch group across 1 directory with 6 updates (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4621">#4621</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/cd5e2b9e73eb8bd783ab872bdda71263b941e2d3"><code>cd5e2b9</code></a> fix(sdk): send Anthropic messages to /v1/compress natively so thinking blocks...</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/6c3d7475077b2e09e8cff5ba2352bda4064d86d3"><code>6c3d747</code></a> chore(sdk-typescript): npm metadata and the proxy requirement up front (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4536">#4536</a>)</li> <li><a href="https://github.com/headroomlabs-ai/headroom/commit/f36772805178e3fed1a95313e8211c57a291b953"><code>f367728</code></a> deps: Bump the npm-minor-patch group across 3 directories with 9 updates (<a href="https://github.com/headroomlabs-ai/headroom/tree/HEAD/sdk/typescript/issues/4495">#4495</a>)</li> <li>See full diff in <a href="https://github.com/headroomlabs-ai/headroom/commits/v0.43.1/sdk/typescript">compare view</a></li> </ul> </details> <br /> Updates `effect` from 4.0.0-beta.103 to 4.1.2 <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/Effect-TS/effect/releases">effect's releases</a>.</em></p> <blockquote> <h2>effect@4.1.2</h2> <h3>Patch Changes</h3> <ul> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10217">#10217</a> <a href="https://github.com/Effect-TS/effect/commit/52166841177ca39d32cc4c06ac5fcdd660656748"><code>5216684</code></a> Thanks <a href="https://github.com/devangpratap"><code>@devangpratap</code></a>! - Fix <code>ByteSize</code> decimal unit constructors such as <code>megabytes(2.01)</code> throwing or returning the wrong byte count, and reject fractional-byte quantities instead of rounding them.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10261">#10261</a> <a href="https://github.com/Effect-TS/effect/commit/b707ad1bce3cbf1613fd97622901d7e81e2a34c7"><code>b707ad1</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Fix cluster entity defect restart backoff being shared by all entities of a type and never resetting.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10249">#10249</a> <a href="https://github.com/Effect-TS/effect/commit/46f48afc88b4b4df3cc1c470ae01f31cc7f8d77c"><code>46f48af</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Prevent an interrupted cluster request that keeps defecting from stalling message delivery to other entities.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10218">#10218</a> <a href="https://github.com/Effect-TS/effect/commit/92201bcd47f1789590ff884542051e2123fdb529"><code>92201bc</code></a> Thanks <a href="https://github.com/sbking"><code>@sbking</code></a>! - Defer the runner notification for a discarded persisted send made inside a <code>ClusterSchema.WithTransaction</code> handler until the transaction commits. The message was invisible to the runner before the commit, so delivery waited for the next storage poll.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10260">#10260</a> <a href="https://github.com/Effect-TS/effect/commit/cf329f3b694ba8659946ee74dbbb6cdb6e5535de"><code>cf329f3</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Fix <code>ClusterWorkflowEngine</code> turning transient storage failures while saving durable clocks or deferred completions into workflow defects; they are now retried.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10262">#10262</a> <a href="https://github.com/Effect-TS/effect/commit/1d25d78d9639e2725f4ea8d056a77b3bde9afd78"><code>1d25d78</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Fix <code>ClusterWorkflowEngine</code> blocking shard reacquisition indefinitely after a forced shard release interrupted a workflow resume waiting on its active run.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10226">#10226</a> <a href="https://github.com/Effect-TS/effect/commit/d190d973a7fb2ebb71bfaf1a01adbffef1833555"><code>d190d97</code></a> Thanks <a href="https://github.com/tim-smart"><code>@tim-smart</code></a>! - Fix <code>ClusterWorkflowEngine</code> letting a runner that lost its shard delete a newer activity result, which could run the activity again and change the workflow's history.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10241">#10241</a> <a href="https://github.com/Effect-TS/effect/commit/d7b3b6bd6be4d96babda9d191538fe25c7124f44"><code>d7b3b6b</code></a> Thanks <a href="https://github.com/Marve10s"><code>@Marve10s</code></a>! - Fix <code>HttpClient.withCookiesRef</code> not sending cookies when the request URL is completed after it, for example by the <code>baseUrl</code> of <code>HttpApiClient</code>. Cookies are now selected for the final URL of each request, including redirect hops.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10211">#10211</a> <a href="https://github.com/Effect-TS/effect/commit/a4787453be8e4f360f0212fb0b25bc4a0e18fc72"><code>a478745</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Keep <code>EventLogServer</code> session auth challenges on their RPC connection, so a challenge is released when its client disconnects instead of being retained by the server after an unauthenticated disconnect.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10221">#10221</a> <a href="https://github.com/Effect-TS/effect/commit/69eaa29dfe5c33547ad12068f65d31b2123194c3"><code>69eaa29</code></a> Thanks <a href="https://github.com/clean99"><code>@clean99</code></a>! - Fix <code>Iterable.filterMapWhile</code> iterators yielding values or pulling from the source after they have returned <code>done</code>.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10233">#10233</a> <a href="https://github.com/Effect-TS/effect/commit/3711ba65343a0932840478aa758a2612be0327a7"><code>3711ba6</code></a> Thanks <a href="https://github.com/clean99"><code>@clean99</code></a>! - Fix <code>Iterable.unfold</code> iterators calling the generating function again after they have returned <code>done</code>.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/9972">#9972</a> <a href="https://github.com/Effect-TS/effect/commit/715c5e4b4528889be2ce30163b130d422e7ab68f"><code>715c5e4</code></a> Thanks <a href="https://github.com/eugene1g"><code>@eugene1g</code></a>! - Add <code>Json5.parse</code> and <code>Jsonc.parse</code> to <code>effect/encoding</code>, returning a <code>Result</code> that fails with an <code>EncodingError</code> on invalid input.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10257">#10257</a> <a href="https://github.com/Effect-TS/effect/commit/b0a1ceebcd053d73dbd4fdf836c4d3471dc18f3c"><code>b0a1cee</code></a> Thanks <a href="https://github.com/tim-smart"><code>@tim-smart</code></a>! - Use a <code>ClientCapabilities</code> instance for the fallback MCP client profile in <code>McpServer</code>.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10242">#10242</a> <a href="https://github.com/Effect-TS/effect/commit/2a10622e906caccd7a9b4a90d1aaef090e8e00cc"><code>2a10622</code></a> Thanks <a href="https://github.com/Marve10s"><code>@Marve10s</code></a>! - Fix <code>McpServer</code> running middleware provided outside a combined middleware inside it.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10243">#10243</a> <a href="https://github.com/Effect-TS/effect/commit/6bccdd54e1bc5d7f182972811d6827d1ad10f1aa"><code>6bccdd5</code></a> Thanks <a href="https://github.com/Marve10s"><code>@Marve10s</code></a>! - Encode declared <code>McpServer</code> tool failures with their JSON codec, so values such as <code>bigint</code>, <code>Map</code> and <code>Option</code> no longer cause an internal error or lossy output</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10244">#10244</a> <a href="https://github.com/Effect-TS/effect/commit/60be01136a530ec10d8b67efc812eee94804444c"><code>60be011</code></a> Thanks <a href="https://github.com/Marve10s"><code>@Marve10s</code></a>! - Fix 2026-07-28 MCP listings failing for resources, resource templates and prompts registered as plain objects.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10235">#10235</a> <a href="https://github.com/Effect-TS/effect/commit/e1edcdc8d60bef515a2c5fa97706d4baaf0c0290"><code>e1edcdc</code></a> Thanks <a href="https://github.com/schutzelaars"><code>@schutzelaars</code></a>! - Stop OTLP log and trace exporters from keeping a timer pending while idle. A Node program kept alive only by an idle exporter now exits.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10210">#10210</a> <a href="https://github.com/Effect-TS/effect/commit/ef81e9dbd1221ab8a9c38ce50f96129054e3b6e2"><code>ef81e9d</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Fix <code>RateLimiter.layerStoreRedis</code> sharing state between different keys. A key ending in <code>:refill</code> or <code>:adaptive</code> no longer reads or writes another key's token-bucket refill time or adaptive state.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10223">#10223</a> <a href="https://github.com/Effect-TS/effect/commit/013fb34f88f3451ede7e7333f205563f07716b13"><code>013fb34</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Fix <code>RpcServer</code> returning an empty HTTP 500 for requests with malformed headers; such requests now fail with a defect <code>Exit</code>, and malformed notifications are dropped.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10251">#10251</a> <a href="https://github.com/Effect-TS/effect/commit/113dbc74b259b7e56e618a506dc67e85794c36ed"><code>113dbc7</code></a> Thanks <a href="https://github.com/wmaurer"><code>@wmaurer</code></a>! - Fix <code>Schedule.fixed</code>, <code>Schedule.windowed</code> and <code>Schedule.cron</code> returning a zero delay when the clock is a fraction of a nanosecond before the next run.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10230">#10230</a> <a href="https://github.com/Effect-TS/effect/commit/519d66eb5972ef3c801fb164dd652f25a71e0511"><code>519d66e</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Add <code>SchemaBinary.int32</code>, <code>uint32</code>, <code>int64</code> and <code>uint64</code>, opt-in schemas that write integers as fixed-width little-endian payloads. The 32-bit schemas decode to <code>number</code> and the 64-bit schemas to <code>bigint</code>. Out-of-range and non-integer values are rejected instead of wrapped.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10213">#10213</a> <a href="https://github.com/Effect-TS/effect/commit/e7a6ea43fe8ac77f68329b84d26dc2d1e136d745"><code>e7a6ea4</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - SQL transactions now trace <code>BEGIN</code>, <code>COMMIT</code>, <code>ROLLBACK</code> and savepoint statements as child spans of <code>sql.transaction</code>, replacing its <code>effect.sql.transaction.commit</code>, <code>savepoint</code> and <code>rollback</code> events; with <code>Statement.SpanPropagationEnabled</code>, the transaction also parents connection spans and records its connection wait time.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10240">#10240</a> <a href="https://github.com/Effect-TS/effect/commit/019f35e0a7ebe050eae01d68cc14ae63158fd8e4"><code>019f35e</code></a> Thanks <a href="https://github.com/clean99"><code>@clean99</code></a>! - Fix <code>Stream.groupAdjacentBy</code> dropping its last group when the stream fails. The group is now emitted before the error.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10278">#10278</a> <a href="https://github.com/Effect-TS/effect/commit/2a0d16cd1ff63a69baf5d19a13dc2051b500150c"><code>2a0d16c</code></a> Thanks <a href="https://github.com/tim-smart"><code>@tim-smart</code></a>! - Fix <code>AnthropicStructuredOutput.toCodecAnthropic</code>, <code>OpenAiStructuredOutput.toCodecOpenAI</code>, and MCP tool schemas returning or rejecting a root <code>$ref</code> when the schema is an identified alias of another identified schema.</p> </li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Changelog</summary> <p><em>Sourced from <a href="https://github.com/Effect-TS/effect/blob/main/packages/effect/CHANGELOG.md">effect's changelog</a>.</em></p> <blockquote> <h2>4.1.2</h2> <h3>Patch Changes</h3> <ul> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10217">#10217</a> <a href="https://github.com/Effect-TS/effect/commit/52166841177ca39d32cc4c06ac5fcdd660656748"><code>5216684</code></a> Thanks <a href="https://github.com/devangpratap"><code>@devangpratap</code></a>! - Fix <code>ByteSize</code> decimal unit constructors such as <code>megabytes(2.01)</code> throwing or returning the wrong byte count, and reject fractional-byte quantities instead of rounding them.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10261">#10261</a> <a href="https://github.com/Effect-TS/effect/commit/b707ad1bce3cbf1613fd97622901d7e81e2a34c7"><code>b707ad1</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Fix cluster entity defect restart backoff being shared by all entities of a type and never resetting.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10249">#10249</a> <a href="https://github.com/Effect-TS/effect/commit/46f48afc88b4b4df3cc1c470ae01f31cc7f8d77c"><code>46f48af</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Prevent an interrupted cluster request that keeps defecting from stalling message delivery to other entities.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10218">#10218</a> <a href="https://github.com/Effect-TS/effect/commit/92201bcd47f1789590ff884542051e2123fdb529"><code>92201bc</code></a> Thanks <a href="https://github.com/sbking"><code>@sbking</code></a>! - Defer the runner notification for a discarded persisted send made inside a <code>ClusterSchema.WithTransaction</code> handler until the transaction commits. The message was invisible to the runner before the commit, so delivery waited for the next storage poll.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10260">#10260</a> <a href="https://github.com/Effect-TS/effect/commit/cf329f3b694ba8659946ee74dbbb6cdb6e5535de"><code>cf329f3</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Fix <code>ClusterWorkflowEngine</code> turning transient storage failures while saving durable clocks or deferred completions into workflow defects; they are now retried.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10262">#10262</a> <a href="https://github.com/Effect-TS/effect/commit/1d25d78d9639e2725f4ea8d056a77b3bde9afd78"><code>1d25d78</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Fix <code>ClusterWorkflowEngine</code> blocking shard reacquisition indefinitely after a forced shard release interrupted a workflow resume waiting on its active run.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10226">#10226</a> <a href="https://github.com/Effect-TS/effect/commit/d190d973a7fb2ebb71bfaf1a01adbffef1833555"><code>d190d97</code></a> Thanks <a href="https://github.com/tim-smart"><code>@tim-smart</code></a>! - Fix <code>ClusterWorkflowEngine</code> letting a runner that lost its shard delete a newer activity result, which could run the activity again and change the workflow's history.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10241">#10241</a> <a href="https://github.com/Effect-TS/effect/commit/d7b3b6bd6be4d96babda9d191538fe25c7124f44"><code>d7b3b6b</code></a> Thanks <a href="https://github.com/Marve10s"><code>@Marve10s</code></a>! - Fix <code>HttpClient.withCookiesRef</code> not sending cookies when the request URL is completed after it, for example by the <code>baseUrl</code> of <code>HttpApiClient</code>. Cookies are now selected for the final URL of each request, including redirect hops.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10211">#10211</a> <a href="https://github.com/Effect-TS/effect/commit/a4787453be8e4f360f0212fb0b25bc4a0e18fc72"><code>a478745</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Keep <code>EventLogServer</code> session auth challenges on their RPC connection, so a challenge is released when its client disconnects instead of being retained by the server after an unauthenticated disconnect.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10221">#10221</a> <a href="https://github.com/Effect-TS/effect/commit/69eaa29dfe5c33547ad12068f65d31b2123194c3"><code>69eaa29</code></a> Thanks <a href="https://github.com/clean99"><code>@clean99</code></a>! - Fix <code>Iterable.filterMapWhile</code> iterators yielding values or pulling from the source after they have returned <code>done</code>.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10233">#10233</a> <a href="https://github.com/Effect-TS/effect/commit/3711ba65343a0932840478aa758a2612be0327a7"><code>3711ba6</code></a> Thanks <a href="https://github.com/clean99"><code>@clean99</code></a>! - Fix <code>Iterable.unfold</code> iterators calling the generating function again after they have returned <code>done</code>.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/9972">#9972</a> <a href="https://github.com/Effect-TS/effect/commit/715c5e4b4528889be2ce30163b130d422e7ab68f"><code>715c5e4</code></a> Thanks <a href="https://github.com/eugene1g"><code>@eugene1g</code></a>! - Add <code>Json5.parse</code> and <code>Jsonc.parse</code> to <code>effect/encoding</code>, returning a <code>Result</code> that fails with an <code>EncodingError</code> on invalid input.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10257">#10257</a> <a href="https://github.com/Effect-TS/effect/commit/b0a1ceebcd053d73dbd4fdf836c4d3471dc18f3c"><code>b0a1cee</code></a> Thanks <a href="https://github.com/tim-smart"><code>@tim-smart</code></a>! - Use a <code>ClientCapabilities</code> instance for the fallback MCP client profile in <code>McpServer</code>.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10242">#10242</a> <a href="https://github.com/Effect-TS/effect/commit/2a10622e906caccd7a9b4a90d1aaef090e8e00cc"><code>2a10622</code></a> Thanks <a href="https://github.com/Marve10s"><code>@Marve10s</code></a>! - Fix <code>McpServer</code> running middleware provided outside a combined middleware inside it.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10243">#10243</a> <a href="https://github.com/Effect-TS/effect/commit/6bccdd54e1bc5d7f182972811d6827d1ad10f1aa"><code>6bccdd5</code></a> Thanks <a href="https://github.com/Marve10s"><code>@Marve10s</code></a>! - Encode declared <code>McpServer</code> tool failures with their JSON codec, so values such as <code>bigint</code>, <code>Map</code> and <code>Option</code> no longer cause an internal error or lossy output</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10244">#10244</a> <a href="https://github.com/Effect-TS/effect/commit/60be01136a530ec10d8b67efc812eee94804444c"><code>60be011</code></a> Thanks <a href="https://github.com/Marve10s"><code>@Marve10s</code></a>! - Fix 2026-07-28 MCP listings failing for resources, resource templates and prompts registered as plain objects.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10235">#10235</a> <a href="https://github.com/Effect-TS/effect/commit/e1edcdc8d60bef515a2c5fa97706d4baaf0c0290"><code>e1edcdc</code></a> Thanks <a href="https://github.com/schutzelaars"><code>@schutzelaars</code></a>! - Stop OTLP log and trace exporters from keeping a timer pending while idle. A Node program kept alive only by an idle exporter now exits.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10210">#10210</a> <a href="https://github.com/Effect-TS/effect/commit/ef81e9dbd1221ab8a9c38ce50f96129054e3b6e2"><code>ef81e9d</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Fix <code>RateLimiter.layerStoreRedis</code> sharing state between different keys. A key ending in <code>:refill</code> or <code>:adaptive</code> no longer reads or writes another key's token-bucket refill time or adaptive state.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10223">#10223</a> <a href="https://github.com/Effect-TS/effect/commit/013fb34f88f3451ede7e7333f205563f07716b13"><code>013fb34</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Fix <code>RpcServer</code> returning an empty HTTP 500 for requests with malformed headers; such requests now fail with a defect <code>Exit</code>, and malformed notifications are dropped.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10251">#10251</a> <a href="https://github.com/Effect-TS/effect/commit/113dbc74b259b7e56e618a506dc67e85794c36ed"><code>113dbc7</code></a> Thanks <a href="https://github.com/wmaurer"><code>@wmaurer</code></a>! - Fix <code>Schedule.fixed</code>, <code>Schedule.windowed</code> and <code>Schedule.cron</code> returning a zero delay when the clock is a fraction of a nanosecond before the next run.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10230">#10230</a> <a href="https://github.com/Effect-TS/effect/commit/519d66eb5972ef3c801fb164dd652f25a71e0511"><code>519d66e</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - Add <code>SchemaBinary.int32</code>, <code>uint32</code>, <code>int64</code> and <code>uint64</code>, opt-in schemas that write integers as fixed-width little-endian payloads. The 32-bit schemas decode to <code>number</code> and the 64-bit schemas to <code>bigint</code>. Out-of-range and non-integer values are rejected instead of wrapped.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10213">#10213</a> <a href="https://github.com/Effect-TS/effect/commit/e7a6ea43fe8ac77f68329b84d26dc2d1e136d745"><code>e7a6ea4</code></a> Thanks <a href="https://github.com/effect-bot"><code>@effect-bot</code></a>! - SQL transactions now trace <code>BEGIN</code>, <code>COMMIT</code>, <code>ROLLBACK</code> and savepoint statements as child spans of <code>sql.transaction</code>, replacing its <code>effect.sql.transaction.commit</code>, <code>savepoint</code> and <code>rollback</code> events; with <code>Statement.SpanPropagationEnabled</code>, the transaction also parents connection spans and records its connection wait time.</p> </li> <li> <p><a href="https://redirect.github.com/Effect-TS/effect/pull/10240">#10240</a> <a href="https://github.com/Effect-TS/effect/commit/019f35e0a7ebe050eae01d68cc14ae63158fd8e4"><code>019f35e</code></a> Thanks <a href="https://github.com/clean99"><code>@clean99</code></a>! - Fix <code>Stream.groupAdjacentBy</code> dropping its last group when the stream fails. The group is now emitted before the error.</p> </li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/Effect-TS/effect/commit/506eef143bcc1750ce58920aac1e45da7265d103"><code>506eef1</code></a> Version Packages (<a href="https://github.com/Effect-TS/effect/tree/HEAD/packages/effect/issues/10212">#10212</a>)</li> <li><a href="https://github.com/Effect-TS/effect/commit/b0a1ceebcd053d73dbd4fdf836c4d3471dc18f3c"><code>b0a1cee</code></a> Update repository dependencies (<a href="https://github.com/Effect-TS/effect/tree/HEAD/packages/effect/issues/10257">#10257</a>)</li> <li><a href="https://github.com/Effect-TS/effect/commit/2a4ce2e3638e5543f1355ae671bcb20e00f45ffe"><code>2a4ce2e</code></a> Make load-sensitive Data, sql-d1 and sqlite-node tests robust under load (<a href="https://github.com/Effect-TS/effect/tree/HEAD/packages/effect/issues/10">#10</a>...</li> <li><a href="https://github.com/Effect-TS/effect/commit/e5b1bcfa48a9e4d0d78de255f5ff3b0a7e2e9727"><code>e5b1bcf</code></a> Consolidate redundant Data, Chunk.getUnsafe and Metric.track tests (<a href="https://github.com/Effect-TS/effect/tree/HEAD/packages/effect/issues/10280">#10280</a>)</li> <li><a href="https://github.com/Effect-TS/effect/commit/4a4491d60c9db1cf43a2346fc49329961512f849"><code>4a4491d</code></a> Fix flaky DurableDeferred.raceAll unprocessed-result replay test (<a href="https://github.com/Effect-TS/effect/tree/HEAD/packages/effect/issues/10265">#10265</a>)</li> <li><a href="https://github.com/Effect-TS/effect/commit/7a0da6300449d4806a25cbd1dde86dc1ef3cf300"><code>7a0da63</code></a> Report the stuck step when an McpServer list change test hangs (<a href="https://github.com/Effect-TS/effect/tree/HEAD/packages/effect/issues/10275">#10275</a>)</li> <li><a href="https://github.com/Effect-TS/effect/commit/2a0d16cd1ff63a69baf5d19a13dc2051b500150c"><code>2a0d16c</code></a> Follow chained root references in structured-output schema conversion (<a href="https://github.com/Effect-TS/effect/tree/HEAD/packages/effect/issues/10278">#10278</a>)</li> <li><a href="https://github.com/Effect-TS/effect/commit/e1edcdc8d60bef515a2c5fa97706d4baaf0c0290"><code>e1edcdc</code></a> Stop idle OTLP log and trace exporters from keeping a timer pending (<a href="https://github.com/Effect-TS/effect/tree/HEAD/packages/effect/issues/10235">#10235</a>)</li> <li><a href="https://github.com/Effect-TS/effect/commit/cf329f3b694ba8659946ee74dbbb6cdb6e5535de"><code>cf329f3</code></a> Retry transient persistence failures for durable clocks and deferreds (<a href="https://github.com/Effect-TS/effect/tree/HEAD/packages/effect/issues/10260">#10260</a>)</li> <li><a href="https://github.com/Effect-TS/effect/commit/46f48afc88b4b4df3cc1c470ae01f31cc7f8d77c"><code>46f48af</code></a> Prevent interrupted defecting cluster requests from stalling other entities (...</li> <li>Additional commits viewable in <a href="https://github.com/Effect-TS/effect/commits/effect@4.1.2/packages/effect">compare view</a></li> </ul> </details> <br /> Updates `@opencode/plugin` from 2.2.5 to 2.2.6 <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/anomalyco/opencode/releases">@opencode/plugin's releases</a>.</em></p> <blockquote> <h2>v2.2.6</h2> <h3>Highlights</h3> <ul> <li><strong>Shared MCP Connections</strong>: Turn on pooling for an MCP server to share one connection across projects with the same setup, so you run fewer processes and start faster. (<a href="https://redirect.github.com/anomalyco/opencode/pull/61185">#61185</a>)</li> <li><strong>GUI Plugin Themes and New Slots</strong>: GUI plugins can now add color themes, render or add actions to code blocks, and use native settings and menu components. (<a href="https://redirect.github.com/anomalyco/opencode/pull/61485">#61485</a>, <a href="https://redirect.github.com/anomalyco/opencode/pull/61533">#61533</a>, <a href="https://redirect.github.com/anomalyco/opencode/pull/61527">#61527</a>, <a href="https://redirect.github.com/anomalyco/opencode/pull/61509">#61509</a>)</li> <li><strong>Mobile Composer Dock</strong>: On phones and tablets, composer actions now sit in a touch dock above the keyboard with larger buttons. (<a href="https://redirect.github.com/anomalyco/opencode/pull/61773">#61773</a>)</li> <li><strong>S... _Description has been truncated_ ## Real Behavior Proof - Fix pushed: migrated `dependency-compat.test.ts` to `effect/cli` `Primitive.FileParse` (module path change in Effect 4.x). - Observed: `bun run typecheck` clean; plugin suite 96/96 pass incl. the compat test (3/3). - Assessment of `toml@0.0.5`: not imported by repo code — the compat test exercises Effect's own TOML parsing, so the dep can be dropped in a follow-up if confirmed unused. - Not tested: other plugins' Effect API surface (only the opencode plugin touches effect/cli). --------- Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: JerrettDavis <mxjerrett@gmail.com>
Headroom (headroom-ai) for TypeScript
Compress tool outputs, logs, JSON and chat history before each LLM call. Fewer input tokens, same answers.
Requires the Headroom proxy. Compression runs in the Python proxy; this
package sends messages to it over HTTP (default http://localhost:8787). Start
it first:
pip install "headroom-ai[proxy]" # or: uv tool install --python 3.13 "headroom-ai[proxy]"
headroom proxy --port 8787
Install
npm install headroom-ai
Quick Start
import { compress } from 'headroom-ai';
const result = await compress(messages, { model: 'gpt-4o' });
console.log(`Saved ${result.tokensSaved} tokens (${((1 - result.compressionRatio) * 100).toFixed(0)}%)`);
// Use compressed messages with any LLM client
const response = await openai.chat.completions.create({
model: 'gpt-4o',
messages: result.messages,
});
Pass baseUrl if the proxy is not on http://localhost:8787.
Framework Adapters
Vercel AI SDK
import { withHeadroom } from 'headroom-ai/vercel-ai';
import { openai } from '@ai-sdk/openai';
import { generateText } from 'ai';
const model = withHeadroom(openai('gpt-4o'));
const { text } = await generateText({ model, messages });
Advanced: using middleware directly
import { headroomMiddleware } from 'headroom-ai/vercel-ai';
import { wrapLanguageModel } from 'ai';
const model = wrapLanguageModel({
model: openai('gpt-4o'),
middleware: headroomMiddleware({ baseUrl: 'http://localhost:8787' }),
});
OpenAI SDK
import { withHeadroom } from 'headroom-ai/openai';
import OpenAI from 'openai';
const client = withHeadroom(new OpenAI());
const response = await client.chat.completions.create({
model: 'gpt-4o',
messages: longConversation,
});
Anthropic SDK
import { withHeadroom } from 'headroom-ai/anthropic';
import Anthropic from '@anthropic-ai/sdk';
const client = withHeadroom(new Anthropic());
const response = await client.messages.create({
model: 'claude-sonnet-4-5-20250929',
messages: longConversation,
max_tokens: 1024,
});
Google Gemini
import { withHeadroom } from 'headroom-ai/gemini';
import { GoogleGenerativeAI } from '@google/generative-ai';
const genAI = new GoogleGenerativeAI(process.env.GEMINI_API_KEY!);
const model = withHeadroom(genAI.getGenerativeModel({ model: 'gemini-2.0-flash' }));
const result = await model.generateContent({
contents: longConversation,
});
HeadroomClient
The full client provides direct access to the proxy's OpenAI and Anthropic passthrough endpoints, plus metrics, CCR, and observability.
import { HeadroomClient } from 'headroom-ai';
const client = new HeadroomClient({
baseUrl: 'http://localhost:8787',
providerApiKey: process.env.OPENAI_API_KEY,
config: {
smartCrusher: { enabled: true, maxItemsAfterCrush: 10 },
ccr: { enabled: true },
},
});
Chat Completions (OpenAI-style)
const response = await client.chat.completions.create({
model: 'gpt-4o',
messages: longConversation,
headroomMode: 'optimize',
});
Messages (Anthropic-style)
const response = await client.messages.create({
model: 'claude-sonnet-4-5-20250929',
messages: longConversation,
max_tokens: 1024,
headroomMode: 'optimize',
});
Direct Compression
const result = await client.compress(messages, { model: 'gpt-4o', tokenBudget: 4000 });
Simulation (Dry Run)
See what compression would do without calling the LLM.
import { simulate } from 'headroom-ai';
const sim = await simulate(messages, { model: 'gpt-4o' });
console.log(`Would save ${sim.tokensSaved} tokens (${sim.estimatedSavings})`);
console.log('Transforms:', sim.transforms);
console.log('Waste signals:', sim.wasteSignals);
console.log('Cache alignment:', sim.cacheAlignmentScore);
Also available on the client:
const sim = await client.chat.completions.simulate({
model: 'gpt-4o',
messages,
});
Compression Hooks
Customize compression with pre/post hooks — matching the Python CompressionHooks API.
import { compress, CompressionHooks } from 'headroom-ai';
import type { CompressContext, CompressEvent } from 'headroom-ai';
class MyHooks extends CompressionHooks {
// Modify messages before compression
preCompress(messages: any[], ctx: CompressContext) {
return [{ role: 'system', content: 'Always preserve error details.' }, ...messages];
}
// Set per-message importance biases
computeBiases(messages: any[], ctx: CompressContext) {
return { 0: 2.0 }; // preserve first message
}
// Observe compression results
postCompress(event: CompressEvent) {
console.log(`Saved ${event.tokensSaved} tokens via ${event.transformsApplied.join(', ')}`);
}
}
const result = await compress(messages, { model: 'gpt-4o', hooks: new MyHooks() });
SharedContext (Multi-Agent)
Compressed inter-agent context sharing — matching the Python SharedContext API.
import { SharedContext } from 'headroom-ai';
const ctx = new SharedContext({ model: 'gpt-4o', ttl: 3600, maxEntries: 100 });
// Agent A stores data (automatically compressed)
const entry = await ctx.put('research', bigAgentOutput, { agent: 'researcher' });
console.log(`Compressed: ${entry.savingsPercent.toFixed(0)}% savings`);
// Agent B reads it (~80% smaller)
const summary = ctx.get('research');
// Agent B gets original if needed
const full = ctx.get('research', { full: true });
// Stats
const stats = ctx.stats();
console.log(`${stats.entries} entries, ${stats.totalTokensSaved} tokens saved`);
CCR Retrieve (Compress-Cache-Retrieve)
Retrieve original content when the LLM needs full details.
const result = await client.compress(messages, { model: 'gpt-4o' });
// Later, when the LLM calls headroom_retrieve:
for (const hash of result.ccrHashes) {
const original = await client.retrieve(hash);
console.log(`${original.originalTokens} original tokens for ${original.toolName}`);
}
// Search within compressed content
const search = await client.retrieve('abc123', { query: 'error logs' });
// Handle LLM tool calls in an agent loop
const toolResult = await client.handleToolCall({
toolCall: assistantMessage.tool_calls[0],
provider: 'openai',
});
Metrics & Observability
// Proxy health
const health = await client.health();
// → { status: 'healthy', version: '0.5.18', config: { optimize: true, ... } }
// Proxy stats
const stats = await client.proxyStats();
// → { requests: { total, cached, failed }, tokens: { saved, savingsPercent }, ... }
// Request metrics
const metrics = await client.getMetrics({ model: 'gpt-4o', limit: 10 });
// Summary
const summary = await client.getSummary();
// Validate setup
const validation = await client.validateSetup();
// Clear cache
await client.clearCache();
// Prometheus metrics
const prom = await client.prometheusMetrics();
Telemetry, Feedback & TOIN
Access the proxy's learning systems.
// Telemetry
const telemetry = await client.telemetry.getStats();
const tools = await client.telemetry.getTools();
// Feedback — per-tool compression hints
const hints = await client.feedback.getHints('list_servers');
// → { hints: { maxItems: 8, skipCompression: false, preserveFields: ['id', 'status'] } }
// TOIN (Tool Output Intelligence Network)
const toinStats = await client.toin.getStats();
const patterns = await client.toin.getPatterns(20);
const pattern = patterns[0] && await client.toin.getPattern(patterns[0].hash);
Each hash is the complete scoped aggregation key, including tenant, auth mode,
model family, and tool signature. Pass it back unchanged for an exact lookup;
the client URL-encodes it. Shorter aggregation-key prefixes are accepted only
when they identify one pattern; ambiguous prefixes return HTTP 409 rather than
an arbitrary pattern.
Configuration Types
Full TypeScript interfaces for every Python config dataclass.
import type { HeadroomConfig, SmartCrusherConfig, CCRConfig } from 'headroom-ai';
const config: HeadroomConfig = {
defaultMode: 'optimize',
smartCrusher: {
enabled: true,
minItemsToAnalyze: 5,
maxItemsAfterCrush: 10,
varianceThreshold: 2.0,
relevance: { tier: 'hybrid', relevanceThreshold: 0.25 },
anchor: { anchorBudgetPct: 0.25 },
},
ccr: { enabled: true, injectTool: true },
cacheOptimizer: { enabled: true, autoDetectProvider: true },
intelligentContext: { enabled: true, useImportanceScoring: true },
};
const client = new HeadroomClient({ config });
Error Handling
Full error hierarchy matching the Python SDK.
import {
HeadroomError,
HeadroomConnectionError,
HeadroomAuthError,
HeadroomCompressError,
ConfigurationError,
ProviderError,
StorageError,
TokenizationError,
CacheError,
ValidationError,
TransformError,
} from 'headroom-ai';
try {
await client.compress(messages);
} catch (err) {
if (err instanceof HeadroomAuthError) {
console.error('Auth failed — check HEADROOM_API_KEY');
} else if (err instanceof HeadroomCompressError) {
console.error(`Compression error ${err.statusCode}: ${err.errorType}`);
} else if (err instanceof ConfigurationError) {
console.error('Bad config:', err.details);
}
}
Format Detection & Conversion
Auto-detects and converts between OpenAI, Anthropic, Vercel AI SDK, and Gemini formats.
import { detectFormat, toOpenAI, fromOpenAI } from 'headroom-ai';
const format = detectFormat(messages); // 'openai' | 'anthropic' | 'vercel' | 'gemini'
const openaiMessages = toOpenAI(messages);
const back = fromOpenAI(openaiMessages, format);
The compress() function handles this automatically — pass any format and get the same format back.
Configuration
import { compress } from 'headroom-ai';
const result = await compress(messages, {
model: 'gpt-4o',
baseUrl: 'http://localhost:8787', // proxy URL
apiKey: 'your-api-key', // optional, for authenticated endpoints
timeout: 30000, // ms
fallback: true, // return uncompressed if proxy is down (default)
retries: 1, // retry on transient failures (default)
tokenBudget: 4000, // compress to fit this limit
hooks: new MyHooks(), // pre/post compression hooks
format: 'vercel', // declare the input format; skips autodetection
});
Or use environment variables:
HEADROOM_BASE_URL— proxy URLHEADROOM_API_KEY— optional API key for authenticated endpoints
Utilities
// Case conversion for proxy communication
import { deepCamelCase, deepSnakeCase } from 'headroom-ai';
const tsObj = deepCamelCase({ tokens_before: 100 }); // { tokensBefore: 100 }
const pyObj = deepSnakeCase({ tokensBefore: 100 }); // { tokens_before: 100 }
// SSE stream parsing
import { parseSSE, collectStream } from 'headroom-ai';
// Hook helpers
import { extractUserQuery, countTurns, extractToolCalls } from 'headroom-ai';
Building from source
Use Node 24 LTS at version 24.11 or later for SDK development. The build
toolchain also supports Node 22.18 or later in the Node 22 line, and Node 26
or later. These are the tsup 9 build requirements; the SDK and optional
provider peers retain their own runtime requirements.
npm ci
npm run typecheck
npm test
npm run build
License
Apache-2.0