Robin/openwiki-wiki
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
ebb95c0a85b85e5abb8d11463cfa8b855c504e5f
openwiki-wiki/sources/README.md
T

43 lines
19 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
---
type: "Reference"
title: "Sources"
description: "Per-connector evidence indexes and ingestion coverage for the local knowledge wiki; durable synthesis lives on canonical pages."
tags: [sources, connectors, evidence, ingestion, 2026-08-25, 2026-08-24, 2026-08-23]
timestamp: 2026-08-25T18:13:00Z
---
# Sources
Per-connector evidence indexes. Each connector gets its own page (for example `gmail.md`, `slack.md`) that records ingestion coverage, freshness, and a compact evidence trail. Durable synthesis lives on the canonical pages (`/themes.md`, `/commitments.md`, `/personal-logistics.md`, `/open-questions.md`); these source pages are the evidence layer underneath.
## Active connectors
| Connector | Page | Status |
| --- | --- | --- |
| Hacker News (`hackernews-1`, LLM-agent topics) | [/sources/hackernews.md](/sources/hackernews.md) | active (last 2026-08-25 18:00 UTC — fifteenth pull since 2026-08-07 with an empty / noise-only Algolia lane on vendor queries; the 2026-08-25 pull is a fully-zero pull (organoid 0 + CelVivo 0 + 3Brain 0 + Cellink 0 + openwiki 0 + deepagents 0, `nbHits=null` in the runtime payload) — **2026-08-23 + 2026-08-24 + 2026-08-25 are three consecutive fully-zero Algolia lanes**, the **longest fully-zero Algolia streak in the 2026-08-07 streak to date**, bookending the 2026-08-22 noise-only pull; the empty-pulls-on-vendor-queries cadence is normal under the 6-query LLM-agent topic set, not an outage, no config-side intervention needed; the noise-only Algolia substring hits include **1 thematic organoid duplicate** (`49400587` Nautilus "The Remarkable Lives of 5-Year-Old Brain Organoids" 1 pt top, the 3rd neural-organoid HN signal in 11 pulls after the 2026-07-27 Nature biocomputing `49071596` + 2026-08-12 Wired `49266068` + 2026-08-20 Nature `49366519` + Science Alert `49373812`) + **2 agent-memory-architecture adjacents** (`49394827` OzBrain 72 pts top, submitter text explicitly cites "Obsidian vaults with 7k files synced through a VPS, markdown repos behind their own MCP servers" — exact architectural parallel to Robin's OpenWiki + Obsidian vault stack; `49399942` Knowl 1 pt top, open-source MCP-backed agent memory with memory-pruning pattern — same pattern the `agent-memory-architecture` row anchors on); the strongest named-stack-adjacent items are `49398152` "Munder Difflin – Agent harness to run an office of your clones" 192 pts top (agent harness cross-walks to the same agent-as-primary-compute-substrate narrative Robin's `hermes-agent-v0-19-quicksilver` row documents; **5th entry in the now 5-day self-hosted-agentic-substrate cluster** — DeepSeek Harness 2026-08-13 + machine0 2026-08-18 + OneCLI 2026-08-19 + self-hosted agentic software factory 2026-08-21 + **Munder Difflin 2026-08-22** — 5 independent submissions across 5 consecutive working days, with Munder Difflin's 192-pt engagement being the second-highest of the cluster after the 422-pt DeepSeek Harness; cluster is now an emerging-pattern watchlist candidate) + `49399591` "New MCP Roadmap" 120 pts top (official MCP roadmap from the canonical MCP blog at blog.modelcontextprotocol.io — **first MCP-protocol roadmap entry since the 2026-07-28 MCP revision finalization** documented on the `hermes-agent-v0-19-quicksilver` row; **MCP substrate is its own substrate-cluster watchlist candidate** — MCP Memory 2026-08-13 + machine0 MCP support 2026-08-18 + OneCLI MCP connectors 2026-08-19 + **New MCP Roadmap 2026-08-22** — 4 distinct substrate-protocol signals across 4 pulls in 10 days; MCP is the protocol-layer substrate of every named-stack item on Robin's workflow — Hermes Agent / Codex CLI / OpenWiki / RAGFlow / machine0 / OneCLI all depend on MCP), with five additional adjacents flagged for traceability: `49394827` OzBrain 72 pts top (3Brain substring-noise duplicate; direct Obsidian-vault + MCP architectural parallel to Robin's OpenWiki stack; adjacent-not-promoted per scope), `49399942` Knowl 1 pt top (Cellink substring-noise duplicate; open-source MCP-backed agent memory with memory-pruning pattern; adjacent-not-promoted per scope), `49400587` Nautilus 5-Year-Old Brain Organoids 1 pt top (organoid/3Brain substring-noise duplicate; 3rd neural-organoid HN signal in 11 pulls; flagged-for-traceability rather than promoted because 1-pt engagement is below the 5-pt 2026-08-12 Wired + 18-pt 2026-08-20 Nature engagement thresholds that drove prior `class-organoid-3d-bioprinting` Last-seen refreshes), `49401549` Anthropic Claude Code A/B testing reduced effort levels 29 pts top (4th Claude Code traceability entry in 14 days; adjacent-not-promoted per scope — Claude Code is not on Robin's named stack), `49394496` "Three important steps in my maturation process" 199 pts best (Thomas Dullien personal-blog essay; adjacent-not-promoted per scope — off-scope per scope discipline), `49397947` "Embedded AI" 29 pts top (No Starch Press book publication; adjacent-not-promoted per scope — book publication, not a vendor stack change)) |
| Git repo (`git-repo-1`, Obsidian vault + OpenWiki wiki) | [/sources/git-repo.md](/sources/git-repo.md) | active (last 2026-08-24 18:05 UTC — **5 new vault commits** in the 2026-08-23T18:11Z → 2026-08-24T18:05Z window (the largest single-day LDR burst since 2026-07-29: CelVivo App Note 2026 + Aiello & Bagni 2025 biorxiv Rotating Chamber + Tian 2025 Nat Commun epicardium mTOR + Kim 2025 Nat Commun EGaIn 3D MEA + Ajongbolo 2025 Cells brain organoid review — all 5 at `04-Literature/Deep-Reads/` root, non-standard pattern); vault HEAD bumped `a4fa5a32` → `fb017e4f`; 2 new automated wiki commits since 2026-08-23T18:11Z fetch (`9cc5e63` daily update 2026-08-24 + `127c2dc` sources-sync 2026-08-24, the **tenth** consecutive sources-sync re-applying the 2026-08-15 count-data wipe pattern after `997b648` 2026-08-16 + `be49285` 2026-08-17 + `1fc73ae` 2026-08-18 + `dcb53a2` 2026-08-19 + `bd30339` 2026-08-20 + `f6c2576` 2026-08-21 + `43d4548` 2026-08-22 + `8f8f767` 2026-08-23 + now `127c2dc` 2026-08-24); 1 new untracked directory `deepread_40240593/` (7 subdirs, LDR pipeline intermediate-artifact workspace)) |
| Web Search (`web-search-1`, Tavily) | [/sources/web-search.md](/sources/web-search.md) | active (latest pull 2026-08-03; 2026-08-08, 2026-08-10, 2026-08-12, 2026-08-15, 2026-08-17, 2026-08-18, 2026-08-19, 2026-08-20, 2026-08-21, 2026-08-22, 2026-08-23, 2026-08-24, and 2026-08-25 update attempts skipped — connector not enabled, `TAVILY_API_KEY` missing; **thirteenth consecutive skip on the same code path**) |
Additional connectors appear here automatically once they are authenticated and ingested at least once.
## Connector roll
| Connector | Status | Auth / env var | Notes |
| --- | --- | --- | --- |
| Gmail (`google`) | not configured | `OPENWIKI_GOOGLE_*` via `openwiki auth google` | Direct API; defaults to `newer_than:1d`. |
| Slack | not configured | `openwiki auth slack` | Needs `search:read` for definitive self-message search; otherwise falls back to recent conversations. |
| Notion | not configured | Notion MCP credentials | Discovery via `openwiki_list_mcp_tools`; do not guess tool names. |
| X / Twitter | not configured | `OPENWIKI_X_ACCESS_TOKEN` via `openwiki auth x` | Streams: `home_timeline`, `user_posts`, `mentions`, `bookmarks`, `list_posts`. |
| Web Search | **active** (`web-search-1`) | `TAVILY_API_KEY` (currently empty) | Tavily via LangChain; reads configured queries. Latest successful pull 2026-08-03 18:08 UTC; the 2026-08-08, 2026-08-10, 2026-08-12, 2026-08-15, 2026-08-17, 2026-08-18, 2026-08-19, 2026-08-20, 2026-08-21, 2026-08-22, 2026-08-23, 2026-08-24, and 2026-08-25 update attempts were skipped because the connector was not enabled in the runtime registry (**thirteenth consecutive skip on the same code path**; nightly searxng-fallback slot for 2026-08-24 is also empty on disk). Evidence at [/sources/web-search.md](/sources/web-search.md). |
| Hacker News | **active** (`hackernews-1`) | _(public, no auth)_ | Public feeds (`top`, `new`, `best`) + Algolia HN search queries. Source-specific query set: `organoid`, `CelVivo`, `3Brain`, `Cellink`, `openwiki`, `deepagents` (LLM-agent topics per the source-specific instruction "Track organoid / CelVivo / 3Brain / Cellink / openwiki / AI agent stories"). Last successful pull 2026-08-25 18:00 UTC — **fifteenth pull since 2026-08-07 with an empty / noise-only Algolia lane on vendor queries** (after the 2026-08-08 03:15 + 2026-08-08 18:00 + 2026-08-13 + 2026-08-15 + 2026-08-16 + 2026-08-18 + 2026-08-19 + 2026-08-21 + 2026-08-23 + 2026-08-24 + 2026-08-25 fully-zero pulls + the 2026-08-10 + 2026-08-12 + 2026-08-17 + 2026-08-20 + 2026-08-22 noise-only pulls; the **2026-08-25 pull adds to 2026-08-23 + 2026-08-24 to form three consecutive fully-zero Algolia lanes**, the **longest fully-zero Algolia streak in the 2026-08-07 streak to date**, bookending the 2026-08-22 noise-only pull; empty-pulls-on-vendor-queries is normal cadence, not an outage, no config-side intervention needed). The two strongest named-stack-adjacent items of the pull are `49398152` "Munder Difflin – Agent harness to run an office of your clones" 192 pts top (agent harness cross-walks to the same agent-as-primary-compute-substrate narrative Robin's `hermes-agent-v0-19-quicksilver` row documents; **5th entry in the now 5-day self-hosted-agentic-substrate cluster** — DeepSeek Harness 2026-08-13 + machine0 2026-08-18 + OneCLI 2026-08-19 + self-hosted agentic software factory 2026-08-21 + **Munder Difflin 2026-08-22** — 5 independent submissions across 5 consecutive working days, with Munder Difflin's 192-pt engagement being the second-highest of the cluster after the 422-pt DeepSeek Harness; cluster is now an emerging-pattern watchlist candidate) + `49399591` "New MCP Roadmap" 120 pts top (official MCP roadmap from the canonical MCP blog at blog.modelcontextprotocol.io; **first MCP-protocol roadmap entry since the 2026-07-28 MCP revision finalization**; **MCP substrate is its own substrate-cluster watchlist candidate** — MCP Memory 2026-08-13 + machine0 MCP support 2026-08-18 + OneCLI MCP connectors 2026-08-19 + **New MCP Roadmap 2026-08-22** — 4 distinct substrate-protocol signals across 4 pulls in 10 days); five additional adjacents flagged for traceability: `49394827` OzBrain 72 pts top (3Brain substring-noise duplicate; direct Obsidian-vault + MCP architectural parallel to Robin's OpenWiki stack; adjacent-not-promoted per scope), `49399942` Knowl 1 pt top (Cellink substring-noise duplicate; open-source MCP-backed agent memory with memory-pruning pattern; adjacent-not-promoted per scope), `49400587` Nautilus 5-Year-Old Brain Organoids 1 pt top (organoid/3Brain substring-noise duplicate; 3rd neural-organoid HN signal in 11 pulls; flagged-for-traceability rather than promoted because 1-pt engagement is below the 5-pt Wired + 18-pt Nature engagement thresholds that drove prior `class-organoid-3d-bioprinting` Last-seen refreshes), `49401549` Anthropic Claude Code A/B testing reduced effort levels 29 pts top (4th Claude Code traceability entry in 14 days; adjacent-not-promoted per scope — Claude Code is not on Robin's named stack), `49394496` "Three important steps in my maturation process" 199 pts best (Thomas Dullien personal-blog essay; adjacent-not-promoted per scope — off-scope per scope discipline), `49397947` "Embedded AI" 29 pts top (No Starch Press book publication; adjacent-not-promoted per scope — book publication, not a vendor stack change). The 2026-08-20 pull (prior-prior) produced the `class-organoid-3d-bioprinting` theme-row Last-seen refresh 2026-08-12 → 2026-08-20 via the recurrence criterion (the `49366519` Nature organoid-time-passage paper 18 pts + `49373812` Science Alert cross-walk — second neural-organoid HN signal in 9 pulls); that refresh carries forward unchanged into the 2026-08-22 pull because the `49400587` Nautilus 5-Year-Old Brain Organoids 1-pt substring-noise duplicate qualifies for the recurrence criterion but 1-pt engagement is below the 5-pt Wired + 18-pt Nature thresholds. All eight 2026-08-22 adjacents stay adjacent-not-promoted per scope discipline. The 2026-08-23 pull (prior) added three named-stack-adjacent items — `49409092` "What Is a Harness?" 121 pts top (6th entry in self-hosted-agentic-substrate cluster, adjacent-not-promoted per scope — definitional essay, not a vendor-stack change), `49402232` "Why your local LLM feels dumber than it is" 450 pts top+best (highest-engagement item of the pull, direct local-LLM-stack cross-walk to the `MiniMax-m3` row, adjacent-not-promoted per scope — third-party local-LLM optimization essay), `49410097` GLM-5.3 vs Anthropic/OpenAI 141 pts top+best (open-weight-LLM-adjacent, cross-walks to the `MiniMax-m3` row's open-weight comparison anchor, adjacent-not-promoted per scope — GLM is not on Robin's named stack), plus `49409073` GLM-5.3 tablet experiment 268 pts top+best + `49407507` Qwen 3.8 27B reverse-engineering 234 pts top+best + `49410149` Vit D + sleep-disturbance/cognition 15 pts top (off-scope) + `49410628` AI agents legal personhood 7 pts top (FT policy opinion, off-scope). The 2026-08-24 pull (prior) added three named-stack-adjacent items — `49411102` FT Anthropic pricing-pressure 738 pts top+best (highest-engagement item of the pull, AI-vendor-business-adjacent cross-walks to the `MiniMax-m3` row's open-weight competitive set pressuring closed-weight pricing, adjacent-not-promoted per scope), `49421554` "Coding expertise is going to collapse from AI reliance" 202 pts top (AI-coding-agent-adjacent, adjacent-not-promoted per scope — coding-talent-debate essay), `49415893` FDA Alzheimer's blood test clearance 154 pts top (only clinical-research-adjacent item, adjacent-not-promoted per scope — Alzheimer's-diagnostic is not one of Robin's 4 application domains). The **2026-08-25 pull** (this run) flagged three named-stack-adjacent items for traceability — `49432317` "Qwen 3.8-Flash-Next releasing tomorrow (125B a6B)" 244 pts top (ModelScope — Alibaba's hosted model hub; **strongest open-weight-LLM-adjacent item of the pull by inference-lane impact**; **seventh open-weight-LLM-adjacent item in the 2026-08-07 streak** — DeepSeek V4 Flash vision 379 pts 2026-08-21 + Unsloth Dynamic 3.0 GGUFs 314 pts 2026-08-19 + GLM-5.3 vs Anthropic/OpenAI 141 pts 2026-08-23 + Qwen 3.8 27B reverse-engineering 234 pts 2026-08-23 + GLM-5.3 tablet 268 pts 2026-08-23 + **Qwen 3.8-Flash-Next 244 pts 2026-08-25** — **1,580+ pts across 6 top-feed items in 6 pulls over 8 days**, the **strongest sustained engagement cluster in the 2026-08-07 streak**; the **Qwen 3.8-Flash-Next "125B a6B" framing** cross-walks to the `MiniMax-m3` row's open-weight comparison anchor (DeepSeek V4 Flash / GLM / Qwen / Llama family); adjacent-not-promoted per scope — Qwen is not on Robin's named stack), `49434291` "Anthropic tells staff to work from home due to possible security team strike" 89 pts top (Business Insider; **strongest AI-vendor-business-adjacent item of the pull by Anthropic operational-distress context** — the **first HN item since the 2026-08-13 `DeepSeek Harness` (422 pts)** to surface a specific Anthropic operational risk; cross-walks to the broader Anthropic-competitive-pressure narrative that the 2026-08-24 FT Anthropic pricing-pressure essay (738 pts) already documented; adjacent-not-promoted per scope — Anthropic operational distress is upstream-of-MiniMax competitive context, not a vendor-stack change Robin acts on), `49437152` "Water Behind the Watts: The Hidden Risk of Powering Data Centers" 15 pts top (Ceres sustainability non-profit; **only data-center-energy-adjacent item on the 2026-08-25 top+best feed lane** + **first water-resource framing in the 2026-08-07 streak**; adjacent-not-promoted per scope — 15-pt engagement is well below all refresh thresholds; data-center water-resource sustainability is not on Robin's named daily-driver stack or 4 application domains), plus two secondary adjacents — `49436400` "Run Minecraft in a Windows Sandbox for Computer Use Agents" 6 pts top (cua.ai computer-use-agent how-to, off-scope) + `49435781` "Warnock: Harnessing GPU Geometry Amplification for Vector Graphics" 17 pts top (ACM Digital Library — GPU geometry amplification for vector graphics rendering with "Harnessing" name-collision only, not an agent-harness item). The 2026-08-25 pull also recorded **zero self-hosted-agentic-substrate cluster entries** — the **second consecutive quiet pull in the cluster** since 2026-08-13 (7 cluster items in 14 days with two misses, density dropping from 0.55 to 0.43 entries/day — **cooling density pattern** worth flagging); the **open-weight-LLM cluster** is now the **strongest sustained engagement cluster in the 2026-08-07 streak** with 1,580+ pts across 6 items in 6 pulls over 8 days — the **most-promotable adjacents cluster** to date but stays adjacent-not-promoted per scope (none of the open-weight LLM models is on Robin's named stack). All 2026-08-23 + 2026-08-24 + 2026-08-25 named-stack-adjacent items stay adjacent-not-promoted under the existing scope discipline. Evidence at [/sources/hackernews.md](/sources/hackernews.md). |
| Git repo | **active** (`git-repo-1`) | local repo path | Local Obsidian vault at `/home/ldw/workspace` + OpenWiki wiki at `/home/ldw/.openwiki/wiki` (two configured repos per `~/.openwiki/connectors/git-repo/config.json`). Writes compact manifest; repo itself is source of truth. Evidence at [/sources/git-repo.md](/sources/git-repo.md). Last updated 2026-08-24 18:05 UTC — **5 new vault commits in the 24-hour window** (the largest single-day LDR burst since 2026-07-29: CelVivo App Note 2026 + Aiello & Bagni 2025 biorxiv Rotating Chamber + Tian 2025 Nat Commun epicardium mTOR + Kim 2025 Nat Commun EGaIn 3D MEA + Ajongbolo 2025 Cells brain organoid review — all 5 at `04-Literature/Deep-Reads/` root, non-standard pattern vs the prior per-vendor subfolder discipline); vault HEAD bumped `a4fa5a32` → `fb017e4f`; 2 new automated wiki commits since the 2026-08-23 fetch (`9cc5e63` daily update 2026-08-24 + `127c2dc` sources-sync 2026-08-24, the **tenth** consecutive sources-sync re-applying the 2026-08-15 count-data wipe pattern after `997b648` 2026-08-16 + `be49285` 2026-08-17 + `1fc73ae` 2026-08-18 + `dcb53a2` 2026-08-19 + `bd30339` 2026-08-20 + `f6c2576` 2026-08-21 + `43d4548` 2026-08-22 + `8f8f767` 2026-08-23 + now `127c2dc` 2026-08-24); 1 new untracked directory `deepread_40240593/` (LDR pipeline intermediate-artifact workspace with 7 subdirs: 5 PMID-named `deepread_<PMID>/` + 2 slug-named `deepread_brain_organoid_fluid/` + `deepread_shear_stress_cs_os/` that exactly map to the Aiello + CelVivo App Note papers). |
## How new source pages get created
When a connector ingests for the first time, an update run will:
1. Write the raw dump to `~/.openwiki/connectors/<connector>/raw/`.
2. Create `/sources/<connector>.md` with the ingestion manifest, freshness, and compact evidence notes.
3. Route durable items into the canonical pages and add a row in `/themes.md` if a recurring signal emerges.
Until then, the new connector only has a row in the active connectors table above and no evidence page of its own..its own...above and no evidence page of its own..its own...
Reference in new issue View Git Blame Copy Permalink