Benchmarks with receipts, postmortems, and releases — HybridDB, CoreMem, ModelPanel and the rest of the stack.https://openassistants.org/https://openassistants.org/blog/assistant-v0-6-18-release/https://openassistants.org/blog/assistant-v0-6-18-release/The latest assistant release closes the gap between what a tool did and what governance recorded: explicit outcomes, fail-closed subagents, and real sandbox enforcement for custom commands.Fri, 25 Sep 2026 00:00:00 GMTreleasesassistanthttps://openassistants.org/blog/modelpanel-gateway-v0-7-0-user-curated-registry/https://openassistants.org/blog/modelpanel-gateway-v0-7-0-user-curated-registry/The gateway stops shipping a provider roster. You curate the routes you actually use, with preview, history, tombstones, and a clean empty-state boot.Fri, 25 Sep 2026 00:00:00 GMTreleasesmodelpanelhttps://openassistants.org/blog/assistant-v0-6-7-native-tool-policy/https://openassistants.org/blog/assistant-v0-6-7-native-tool-policy/The wrong way to trim an agent's tools is to list what it can't do. Deployment-native tool policy: all, selected, or none — with exact and glob patterns, env precedence, and a hard ceiling.Tue, 15 Sep 2026 00:00:00 GMTreleasesassistanthttps://openassistants.org/blog/modelpanel-gateway-v0-3-0-release/https://openassistants.org/blog/modelpanel-gateway-v0-3-0-release/Aliases, combos, presets — removed. One way to fuse: a named panel. The breaking change that says what the gateway is, after the release that made it complete.Fri, 11 Sep 2026 00:00:00 GMTreleasesmodelpanelhttps://openassistants.org/blog/assistant-v0-6-3-governance-that-cant-be-bypassed/https://openassistants.org/blog/assistant-v0-6-3-governance-that-cant-be-bypassed/A middleware that vanished, a stream that hung forever when blocked — and the hotfix release that made governance terminal instead of optional.Thu, 10 Sep 2026 00:00:00 GMTreleasesassistanthttps://openassistants.org/blog/agent-memory-benchmarks-are-incomparable/https://openassistants.org/blog/agent-memory-benchmarks-are-incomparable/99.9% vs 96.6% — neither number means what you think. Why recall@5, answer accuracy, and eval sets keep getting conflated, and the full, reproducible measurement behind CoreMem's claims.Tue, 08 Sep 2026 00:00:00 GMTbenchmarkscorememmemoryhttps://openassistants.org/blog/agentprofile-v0-2-1-release/https://openassistants.org/blog/agentprofile-v0-2-1-release/A PROFILE.md that carries its own schema version and hard cost caps — now on PyPI, because identity you can't install is identity you won't use.Tue, 08 Sep 2026 00:00:00 GMTreleasesagentprofilehttps://openassistants.org/blog/failed-experiments-are-benchmarks-too/https://openassistants.org/blog/failed-experiments-are-benchmarks-too/The reranker that looked better, the graph edges we believed in, the embedder swap that was a wash — and the one rule that kept the default pipeline honest: the composition lesson.Tue, 08 Sep 2026 00:00:00 GMTreceiptscorememmethodologyhttps://openassistants.org/blog/hybriddb-080-one-identity-across-the-store/https://openassistants.org/blog/hybriddb-080-one-identity-across-the-store/Chroma vectors are now keyed by the logical primary key, killing a whole bug class at the root: the TEXT-PK journal wedge and the repeated rowid→pk mapping bugs.Tue, 08 Sep 2026 00:00:00 GMTreleaseshybriddbhttps://openassistants.org/blog/hybriddb-vs-the-embedded-stack/https://openassistants.org/blog/hybriddb-vs-the-embedded-stack/sqlite-vec, LanceDB, KùzuDB, DuckDB are each excellent at one thing. HybridDB exists because agent workloads need all five consistent at once. The BEIR and DuckDB numbers behind the claim.Tue, 08 Sep 2026 00:00:00 GMTbenchmarkshybriddbdatabaseshttps://openassistants.org/blog/coremem-v0-16-0-release/https://openassistants.org/blog/coremem-v0-16-0-release/150 noise messages could push your answer out of the retrieval window entirely. The fix was teaching the vector store to filter before it searches.Wed, 02 Sep 2026 00:00:00 GMTreleasescorememhttps://openassistants.org/blog/coremem-v0-16-1-release/https://openassistants.org/blog/coremem-v0-16-1-release/A message stored at 23:00 was excluded by a filter saying 'after 02:00.' The fix is one line. The story is why it took until now to fix.Wed, 02 Sep 2026 00:00:00 GMTreleasescorememhttps://openassistants.org/blog/hybriddb-v0-7-0-release/https://openassistants.org/blog/hybriddb-v0-7-0-release/The week we found a 352× speedup hiding in plain sight, taught the vector store to filter before searching, and made rollback 20× faster — by batching.Wed, 02 Sep 2026 00:00:00 GMTreleaseshybriddbhttps://openassistants.org/blog/coremem-v0-14-0-release/https://openassistants.org/blog/coremem-v0-14-0-release/Every default your memory system ships is a bet. We spent a release cycle measuring ours — then added tamper-evident governance on top of what survived.Mon, 31 Aug 2026 00:00:00 GMTreleasescorememhttps://openassistants.org/blog/hybriddb-v0-6-0-release/https://openassistants.org/blog/hybriddb-v0-6-0-release/Before we could promise 'history you can verify,' we first had to make the journal boring enough to trust. This is that release, and the war that preceded it.Wed, 26 Aug 2026 00:00:00 GMTreleaseshybriddbhttps://openassistants.org/blog/connectkit-v0-1-6-release/https://openassistants.org/blog/connectkit-v0-1-6-release/The v0.1.x series as one story: from a spec that could be read to a runtime where OAuth tokens refresh themselves and tool descriptions are JSON Schema.Sat, 06 Jun 2026 00:00:00 GMTreleasesconnectkit