Your Codebase's News Reader

Yupa ingests tech news from 20+ verified sources, matches concepts to your repos, and generates feature specs — ready to approve and deploy as PRs.

CONCEPT EXTRACTED SCORE: 87

Claude Agent SDK 2.0 — Streaming Tool Calls

New streaming interface allows agents to emit partial tool results during execution, reducing perceived latency for long-running operations.

Python async FastAPI agents

"You read about a new framework. You think 'might be useful someday.' You never integrate it."

01 · DISCOVERY

Noise Everywhere

Hacker News, arXiv, Twitter, Substacks — you skim dozens of sources daily. Most of it isn't relevant to your stack.

02 · TRANSLATION

Manual Research

"Interesting framework" → "How do we actually use this?" requires hours of reading docs, comparing APIs, and checking compatibility.

03 · INTEGRATION

Never Happens

The ticket gets deprioritized. The bookmark collects dust. Your team adopts the tool 6 months later — after your competitor already shipped it.

Four Steps from News to PR

Yupa's intelligence loop runs daily, automatically.

01

Ingest

20+ verified sources scraped daily. Anthropic, DeepMind, Hacker News, arXiv, Y Combinator, engineering Substacks. Structured concept extraction via Claude Haiku.

02

Match

Your repos are fingerprinted — dependencies, frameworks, architecture. Each concept is scored 0-100 against your actual stack. Language mismatches are pre-filtered.

03

Spec

For high-scoring matches, Claude Sonnet generates a grounded PRD with integration steps, dependencies, risks, and discovery keywords. Every claim cites the source article.

04

Execute

Review the spec in your dashboard. Approve it. Yupa creates a feature branch, generates the code, and opens a PR. You merge when ready.

Why Concept Signatures Matter

Most news aggregators give you titles and summaries. Yupa extracts structured concept signatures — machine-readable representations of what a technology does, what stacks it's compatible with, and what problems it solves.

A raw article about a new Python async library becomes a structured object that Yupa can reason about:

{
  "concept_name": "AnyIO 5.0 — Structured Concurrency",
  "category": "Async Framework",
  "core_technologies": ["Python", "asyncio", "trio"],
  "compatible_architectures": ["FastAPI", "ASGI"],
  "problem_solved": "Task groups with automatic cleanup on failure",
  "relevance_score": 82
}

This isn't keyword matching. Claude Haiku reads the full article and extracts the semantic intent — understanding what a tool does and which codebases would benefit from adopting it.

20+ Verified Intelligence Sources

Curated, not firehosed. Each source is verified for signal quality.

Foundation Model Research

  • Anthropic Research Blog
  • Google DeepMind
  • Meta FAIR
  • OpenAI Blog

Papers & Trending Repos

  • Hugging Face Daily Papers
  • arXiv cs.AI / cs.CL
  • GitHub Trending (AI)

Industry Signal

  • Hacker News (AI/LLM)
  • a16z AI Newsletter
  • Sequoia AI Perspectives
  • Y Combinator Launches

Engineering Deep Dives

  • Latent Space
  • Ahead of AI
  • Interconnects

The Matching Problem

Naive keyword matching would tell you every Python library is relevant to your Python repo. That's useless. Yupa's matching is contextual — it understands what your repo actually does.

A new async streaming library scores 87 against your FastAPI backend (directly improves request handling) but only 12 against your React frontend (wrong language family, wrong architecture). The pre-filter catches the language mismatch before Claude Haiku even runs — saving API costs.

Scores above 60 get specs auto-generated. Scores between 30-60 appear in your digest for manual review. Below 30, they're silently filtered. You only see what matters.

Costs Shown Upfront

No surprise charges. You see what every action costs before it runs.

CONCEPT MATCH

~$0.001

Per concept scored against your repo via Claude Haiku. Pre-filtered by language first.

SPEC GENERATION

~$0.05

Per spec generated via Claude Sonnet with grounded citations. One auto-spec per repo per run.

LOCAL FALLBACK

$0

Run specs locally via Ollama + Nemotron. No API costs. Slightly lower quality, full privacy.

Specs That Don't Hallucinate

Generic AI summaries invent plausible-sounding details. Yupa specs use the Anthropic Citations API to ground every claim in the original source article. If the source doesn't mention a detail, Sonnet says "unknown" instead of guessing.

Each spec includes: what changed in the ecosystem, a concrete integration plan with specific files and imports, dependencies to add or remove, risks and unknowns, and discovery keywords for further research. It's a structured PRD, not a paragraph of marketing copy.

Human review at every step. Nothing executes without your approval. Yupa is your autonomous PM, not your autonomous engineer — it researches and recommends, you decide and deploy.

A Day with Yupa

$ yupa ingest --family "frontier-labs" 14 articles ingested from 4 sources 9 concept signatures extracted $ yupa match --repo ghostpeony/browsy 3 matches found 87 Claude Agent SDK 2.0 — Streaming Tool Calls 72 Playwright MCP — Browser Observation Protocol 65 DOM Spatial Indexing via R-Tree $ yupa spec --match 87 Spec generated: "Add streaming tool results to Browsy's agent interface using the new SDK streaming API" → Integration plan: 3 files, 1 new dependency → Citations: 4 references from source article $ yupa execute --match 87 --approve Branch created: feat/streaming-tool-calls PR #42 opened: "Add streaming tool results"

Frequently Asked Questions

Yupa tracks 20+ verified sources across four families: Frontier Labs (Anthropic, Google DeepMind, Meta FAIR, OpenAI), Open-Source AI (Hugging Face Daily Papers, arXiv, GitHub Trending), VC & Accelerators (Hacker News, a16z, Sequoia, Y Combinator), and Engineering Substacks (Latent Space, Ahead of AI, Interconnects). You can also add your own custom sources.
Yupa fingerprints your connected repos — scanning dependencies, frameworks, and architecture. When new concepts are extracted from articles, Claude Haiku scores each concept against each repo on a 0-100 scale, filtering by language compatibility first. Only matches scoring 60+ get auto-generated specs. 30-60 appear in your digest for manual review.
A Yupa spec is a structured PRD generated by Claude Sonnet. It includes: what changed in the ecosystem, a concrete integration plan with specific files and imports, dependencies to add or remove, risks and unknowns, and discovery keywords for further research. Every claim cites the original source article via the Anthropic Citations API — no hallucinations.
Yupa is transparent about costs. Concept matching via Claude Haiku costs approximately $0.001 per match. Spec generation via Claude Sonnet costs approximately $0.05 per spec. You can also use a local LLM via Ollama to eliminate API costs entirely. Yupa is currently in private beta with free early access.