Robin/openwiki-wiki
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

chore: daily update 2026-09-02

Browse Source
This commit is contained in:
OpenWiki Bot committed 2026-09-02 02:13:51 +08:00
1 parent 0890fdbad8
commit c18d383b1c
13 files changed
+75 -1846
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

themes/03-ai-agents.md
+1
View File
@@ -119,6 +119,7 @@ The architecture shows three agent surfaces sharing the OpenWiki memory substrat
- **`agent-memory-architecture` theme row** on [`/themes.md`](/themes.md) — clusters Karpathy LLM Wiki + Google OKF + MCP + verginglabs.com "A Markdown wiki outscored every AI agent memory product we benchmarked" (HN `49156055`, 2 pts, watchlist). 17+ distinct sources on the shared-agent-memory pattern. OpenWiki aligns with this lane (Open Knowledge Format front-matter is the structural backbone).
- **`Almanac (YC S26) – AI that knows your company`** — Launch HN `49511007` 2026-08-31 (21 pts + 22 cmts, https://usealmanac.com/). **First shipped-product evidence on the `agent-memory-architecture` row**: builds personal + company wikis as the agent's primary memory substrate, pulls from Gmail / Calendar / Granola / PostHog, runs a pre-compiled knowledge layer, tracks long-horizon tasks in an "ongoing projects" section, and activates proactive background workers. Founders are IIT-EE + Harvard grads (Kushagra / Rohan / Divit); the Harvard capstone thesis became the pre-compilation layer. Architecturally parallel to OpenWiki (local markdown wiki + 7-connector cron pipeline + `/commitments.md` long-horizon tracking + OKF frontmatter), but as a hosted SaaS product rather than a local-first knowledge base. Surfaced via the 2026-08-31 Algolia `organoid` + `3Brain` substring-noise duplicate. Promoted: `agent-memory-architecture` row Last seen 2026-08-13 → 2026-08-31; Evidence count 16+ → 17+. **Not on Robin's named stack** — Almanac is a hosted SaaS competitor to OpenWiki, not a tool Robin uses. Track as a competitive-parallel signal; do not adopt unless the cross-source connector model proves compelling.
- **`Show HN: Running 104GB Qwen3.8-Flash-Next on 48GB Mac with at ~12 tok/s`** — HN `49524447` 2026-09-01 (36 pts top + 34 cmts, by `carloslfu`, https://github.com/carloslfu/slotstream — Show HN local-LLM deployment of Qwen3.8-Flash-Next on a 48GB Mac via the slotstream project). **12th entry in the open-weight-LLM cluster** of the 2026-08-07 streak; the cluster now totals ~3,354+ pts across 12 items in 10 pulls over 13 days — the **strongest sustained engagement cluster of the entire 2026-08-07 streak**. The Qwen3.8-Flash-Next-on-Mac item corroborates the `MiniMax-m3` row's open-weight comparison anchor + the unsloth/MiniMax-M3-GGUF local-LLM run path; **promoted** — `MiniMax-m3` row description refresh to mention the 12-item cluster; the **0.94 cmts-to-upvote ratio** indicates substantive technical discussion rather than passive upvote stream. **Adjacent to Robin's named stack** — Qwen is not on Robin's named stack (Hermes / MiniMax-M3+H3 / OpenWiki / RAGFlow / NocoDB / SeaFile / Codex CLI are the named agents) but the 48GB-Mac 12 tok/s deployment pattern is relevant to Robin's potential future local-LLM runs alongside unsloth/MiniMax-M3-GGUF.
- **Tavily / Hermes / MiniMax-M3 lane** — Tavily answer text on this lane is **partially fabricated** (e.g. Tavily claimed a "release date of June 1, 2026" for Hermes v0.19 Quicksilver; the actual release is 2026-07-25 per Nous Research). Trust underlying URLs + GitHub release tags; ignore Tavily answer text.
## Cross-cutting relationships