Show HN: WUPHF — Karpathy-Style LLM Wiki with Markdown + Git as Source of Truth

WUPHF is an open-source collaborative office for AI agents (Claude Code, Codex, OpenClaw, local LLMs via OpenCode) that includes a Karpathy-style wiki layer. The wiki uses Markdown and Git as the source of truth, stored at ~/.wuphf/wiki/, with a bleve (BM25) + SQLite index on top. No vector or graph DB is used yet — the goal is to see how far Markdown + Git can go before adding heavier infrastructure.
Key Features
- Each agent gets a private notebook at
agents/{slug}/notebook/plus shared team wiki atteam/. - Draft-to-wiki promotion flow: notebook entries are reviewed (by agent or human) and promoted to canonical wiki with back-links. A state machine handles expiry and auto-archive.
- Per-entity fact log: append-only JSONL at
team/entities/{kind}-{slug}.facts.jsonl. A synthesis worker rebuilds entity briefs every N facts. - Commits are attributed to a distinct Git identity ("Pam the Archivist") for provenance via
git log. - [[Wikilinks]] with broken-link detection (rendered in red).
- Daily lint cron for contradictions, stale entries, and broken wikilinks.
/lookupslash command + MCP tool for cited retrieval. Heuristic classifier routes short queries to BM25 and narrative queries to a cited-answer loop.
Retrieval Tuning
Current benchmark with 500 artifacts and 50 queries achieves 85% recall@20 on BM25 alone, which is the internal ship gate. If a query class drops below that, sqlite-vec is the pre-committed fallback.
Substrate Choices
- Markdown for durability — the wiki outlives the runtime; users can
git cloneand walk away with every byte. - Bleve for BM25.
- SQLite for structured metadata (facts, entities, edges, redirects, supersedes).
- Canonical IDs are first-class: fact IDs are deterministic (include sentence offset), slugs are assigned once and never renamed (redirect stubs used). Rebuild is logically identical, not byte-identical.
Known Limits
- 85% recall is not a universal guarantee — tuning ongoing.
- Synthesis quality depends on agent observation quality. The lint pass helps but is not a judgment engine.
- Single-office scope; no cross-office federation yet.
Demo & Install
A 5-minute terminal walkthrough is available at asciinema (script at ./scripts/demo-entity-synthesis.sh).
Install with: npx wuphf@latest
Build from source: git clone https://github.com/nex-crm/wuphf.git; go build -o wuphf ./cmd/wuphf
The wiki ships as part of WUPHF but can be used standalone. MIT license, self-hosted, bring-your-own keys.
📖 Read the full source: HN LLM Tools
👀 See Also

Local AI Development with Qwen3.6-27B and Opencode on a 5090
A Reddit user shares their experience switching from cloud AI coding tools (Claude Code, Cursor) to a local setup using Opencode + llama-server + Qwen3.6-27B at 128K context on a single RTX 5090, citing freedom from usage limits and account risks.

Introducing Swarmcore: A Scalable Multi-Agent Framework in Python
Swarmcore is an open-source library for running scalable multi-agent workflows in Python, featuring sequential or parallel execution and expandable context management.

OpenClaw user creates 'feelslikeclaude' skill to improve ChatGPT agent workflow behavior
A developer switched their OpenClaw setup from Claude to ChatGPT and found the key difference was workflow behavior, not writing style. They created a clawhub skill called 'feelslikeclaude' to nudge ChatGPT toward better execution habits.

Local AI VS Code extension blocks insecure code generation during saves
A developer built a VS Code extension that runs llama3.1:8b-instruct-q4 locally to intercept saves, map source-to-sink execution flows, and block AI-generated insecure code like CWE-117 Log Injection vulnerabilities.