The daily firehose — everything the toolchain shipped today, already filtered.
// HOW THIS ISSUE IS MADE
We read every release from the 174 tools on our watchlist at the source — GitHub and GitLab release notes, vendor release pages and changelogs, project blogs and feeds, vendor press releases, and the source code behind the tag. Bug-fix-only releases and non-product newsroom noise are dropped; what's left is summarized down to the new capability, how to try it, and any screenshots or videos the release itself published. Every entry links to the sources it was built from.
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.
Claude Code v2.1.205 adds a full /doctor setup checkup, smarter agent view with PR linking, and a new auto-mode safety rule for rm -rf.
└──▷ GET THIS VERSION
$ git clone --branch v2.1.205 https://github.com/anthropics/claude-code.git
# already have the repo? check out this version:$ git checkout v2.1.205
└──▷ TRY IT
Review agent sessions with PR-linked context and human-readable status headlines.
$ claude agents
›Adds an auto mode rule that blocks tampering with session transcript files.
›Adds auto mode prompt before running rm -rf on a variable it can't resolve from context.
›Improves agent view: sessions that edit, merge, comment on, or push to an existing PR now link it in claude agents.
›Improves agent view: rows now show a colored state word and a classifier-written headline instead of raw tool call text, with full status and exact ask for blocked sessions in the peek.
›Background task notifications now explicitly state that no human input has occurred, preventing fabricated in-transcript approvals from being acted on.
+2 moreshow less
›Auto-update binary downloads now stream to disk instead of buffering in memory, cutting peak updater memory usage by roughly 400 MB.
›Reserves the 'Claude Browser' MCP server name (alongside 'Claude Preview'); user-configured MCP servers can no longer register under either name.
└──▷ BREAKING ON UPGRADE
!User-configured MCP servers can no longer register under the name 'Claude Browser' or 'Claude Preview'; servers using either name will be rejected.
OpenHands cloud gains SMTP email service, user login tracking, and a renamed Usage & Monitoring dashboard.
└──▷ GET THIS VERSION
$ git clone --branch cloud-1.43.0 https://github.com/All-Hands-AI/OpenHands.git
# already have the repo? check out this version:$ git checkout cloud-1.43.0
›Adds SMTP email service integration for outbound email delivery.
›Surfaces email-enabled status for both SMTP and Resend email providers in the UI.
›Tracks user login timestamps for visibility into account activity.
›Renames the admin dashboard to 'Usage & Monitoring'.
Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.
CrewAI 1.15.2 adds inline skill definitions, templated Flow action inputs, stream frame protocol, and AgentExecutor feedback handling.
└──▷ GET THIS VERSION
$ git clone --branch 1.15.2 https://github.com/crewAIInc/crewAI.git
# already have the repo? check out this version:$ git checkout 1.15.2
›Supports inline skill definitions, allowing skills to be defined directly without separate files.
Haystack v2.31.0 adds async evaluators, type-preserving routing, YAML frontmatter extraction, and expanded reference range support.
└──▷ GET THIS VERSION
$ git clone --branch v2.31.0 https://github.com/deepset-ai/haystack.git
# already have the repo? check out this version:$ git checkout v2.31.0
└──▷ USE IT
Route a structured dataclass through a pipeline without losing its type — useful when downstream components expect a typed object, not a string.
python
from haystack.components.routers import ConditionalRouter
routes = [
{
"condition": "{{query.intent == 'search'}}",
"output": "query",
"output_name": "search_query",
"output_type": ParsedQuery,
"output_passthrough": True,
},
]
router = ConditionalRouter(routes)
result = router.run(query=ParsedQuery(text="What is Haystack?", intent="search", entities=[]))
assert result["search_query"].intent == "search" # type preserved
Run multiple RAG evaluations concurrently in a FastAPI service without blocking the event loop.
python
from haystack.components.evaluators import FaithfulnessEvaluator
import asyncio
evaluator = FaithfulnessEvaluator()
result = await evaluator.run_async(
questions=["What is Haystack?"],
contexts=[["Haystack is an AI framework."]],
predicted_answers=["Haystack is an AI framework."]
)
›Adds output_passthrough: True field to ConditionalRouter route definitions, bypassing Jinja2 rendering so complex types like dataclasses and Pydantic models are passed through unchanged rather than silently stringified.
›Adds extract_frontmatter=True parameter to MarkdownToDocument; when set, YAML frontmatter is stripped from converted content and stored in Document.meta.
›Adds expand_reference_ranges parameter to AnswerBuilder; when enabled, citation ranges like [6-10] and [1-3,7-9] are expanded to individual document indices in RAG answers (disabled by default).
›Adds document_comparison_field parameter to DocumentNDCGEvaluator, allowing document matching by 'content', 'id', or any 'meta.<key>' field when calculating NDCG scores, consistent with DocumentMAPEvaluator, DocumentMRREvaluator, and DocumentRecallEvaluator.
›Adds native async support via run_async to LLMEvaluator, FaithfulnessEvaluator, and ContextRelevanceEvaluator, enabling concurrent evaluation in async applications like FastAPI or FastMCP without blocking the event loop.
└──▷ BREAKING ON UPGRADE
!DocumentNDCGEvaluator now matches documents by content instead of id by default; existing pipelines may see changed NDCG scores. Pass document_comparison_field="id" to restore the previous behavior.
›Adds GGML_OPENCL_MOE_RAGGED_GRAN={8,16,32} environment variable to control OpenCL MoE prefill GEMM tile-skip granularity (quarter/half/off), defaulting to quarter (8-column skip-groups) across all eight *_f32_ns MoE GEMMs.
›Adds GGML_OPENCL_MOE_RAGGED_FP16=0 opt-out environment variable to disable ragged-tile half-tile skipping for OpenCL MoE FP16 prefill GEMM.
›New OpenCL ragged-tile MoE prefill optimization skips fully-padded expert tiles during GEMM, reducing wasted compute on low tokens-per-expert workloads while remaining numerically identical to the non-skipped path.
llama.cpp b9909 adds timings and progress fields to the /responses API stream endpoint.
└──▷ GET THIS VERSION
$ git clone --branch b9909 https://github.com/ggml-org/llama.cpp.git
# already have the repo? check out this version:$ git checkout b9909
›Adds timings and progress data to the /responses API streaming endpoint, enabling clients to monitor inference performance and completion progress in real time.
Phoenix v17.21.0 adds end-to-end agent turn tracing, a global search palette, and session lifecycle management.
└──▷ GET THIS VERSION
$ git clone --branch arize-phoenix-v17.21.0 https://github.com/Arize-ai/phoenix.git
# already have the repo? check out this version:$ git checkout arize-phoenix-v17.21.0
›Adds end-to-end PXI turn tracing from browser to backend for agents, linking frontend interactions to backend spans in a single trace.
›Adds a global search command palette (⌘K) for fast in-app navigation across projects and traces.
›Adds a session stats side panel to the sessions table, surfacing per-session metrics without leaving the list view.
›Adds session turn dividers with copy support for easier inspection and sharing of individual turns.
›Automatically sweeps orphaned sessions after a one-hour grace period to keep session data clean.
+2 moreshow less
›Shows total session count on project cards for at-a-glance visibility.
›Introduces read-only UI styles for surfaces that should not be editable.
Langfuse v3.208.0 makes Parquet the default export format and tightens prompt label handling via MCP.
└──▷ GET THIS VERSION
$ git clone --branch v3.208.0 https://github.com/langfuse/langfuse.git
# already have the repo? check out this version:$ git checkout v3.208.0
›Parquet is now the default export format for all projects (previously required opt-in per project).
›MCP prompt creation now blocks applying the production label directly at creation time, enforcing a safer promotion workflow.
└──▷ BREAKING ON UPGRADE
!The default blob export format for all projects is now Parquet; any downstream pipeline that expected a different default format will receive Parquet output without explicit configuration.
THE AI TOOLCHAIN