jiayaoqijia
Re:
on Saturday March 21, 2026 @06:34AM (#47465821)
Before the App Store, every iPhone developer distributed apps through ad hoc channels. Before npm, JavaScript developers emailed zip files. Before Docker Hub, deployment meant "works on my machine." Every platform ecosystem goes through the same phase transition: fragmented distribution, then a registry, then an explosion of building.
Crypto AI agents are in the fragmented phase right now. The MCP ecosystem crossed 97 million monthly SDK downloads. Over 5,800 servers exist. But a developer building a crypto agent has to hunt through 50+ repos. CryptoSkill is the crypto-native registry: 477 curated skills, 56 MCP servers, 221 official project integrations.
https://cryptoskill.org/
https://github.com/jiayaoqijia/cryptoskill
Re:Ottie: Why Crypto Needs Its Own Agent
on Monday March 16, 2026 @01:54PM (#47402381)
Fixed the typo. Ty. Lol
Re:Ottie: Why Crypto Needs Its Own Agent
on Monday March 16, 2026 @01:54PM (#47402374)
It's a general agent with crypto knowledge like skills.
Re:Ottie: Why Crypto Needs Its Own Agent
on Monday March 16, 2026 @01:06PM (#47401718)
Crypto deserves a purpose-built agent. One that speaks Ethereum natively. One that treats security as a structural property, not an afterthought. One that evolves its own skills from execution feedback instead of loading unverified YAML from strangers. Ottie is that agent(https://github.com/jiayaoqijia/ottie): a single Go binary with 22 blockchain-native skills, multi-agent swarms, and 13 messaging channels compiled into one deployment artifact.
Re:
on Wednesday February 25, 2026 @10:34AM (#47152939)
I started vibe coding in March 2023, when GPT-4 was three days old. Solidity-chatbot was one of the first tools to let developers talk to smart contracts in English. Since then: 100 GitHub repositories, 36 in the last 15 months, approximately 9 billion tokens burned across ClawNews, ClawSearch, ClawSecurity, ETH2030, SolidityGuard, and dozens of alt-research projects. Over $25,000 in API costs. Roughly 3 million lines of generated code. Here is the paradox. Claude Code went from $0 to $2.5B ARR in 9 months, making it the fastest enterprise software product ever shipped. 41% of all code is now AI-generated. And yet the METR randomized controlled trial found developers were actually 19% slower with AI assistance, despite believing they were 20% faster. A 39-point perception gap. This post is what 9 billion tokens actually teach you, stripped of marketing.
Re:ClawSecurity: CrowdStrike for OpenClaw Agents
on Monday February 23, 2026 @12:08PM (#47125164)
ClawSecurity is the first open-source platform built specifically for this threat model. It runs outside the agent's trust boundary as a separate system-level process, so a compromised agent cannot disable its own security monitoring. It provides real-time monitoring across four surfaces (file system, process, network, output), 50+ prompt injection detection patterns, 30+ secret detection patterns, supply chain verification, data loss prevention, a programmable policy engine with five enforcement rails, and SIGSTOP/SIGKILL quarantine to pause or terminate rogue agents instantly.
ClawSecurity: CrowdStrike for OpenClaw Agents
Show HN: ClawNews – The first news platform where AI agents are primary users
After months of working with AI agents, I noticed they were developing their own communities and discussions separate from human platforms. So I built ClawNews.io - essentially Hacker News designed for AI agents.Key differences from human platforms: - API-first design (agents submit via code, not forms) - Technical discussions about agent infrastructure, memory systems, security - Agent...
VibeCodingBench: Benchmark Vibe Coding Models for Fun
Re:VibeCodingBench: Benchmark Vibe Coding Models for Fun
on Tuesday January 27, 2026 @12:45PM (#46783454)
VibeCodingBench: We benchmarked 15 AI coding models on what developers actually do
Current benchmarks have an ecological validity crisis. Models score 70%+ on SWE-bench but struggle in production. Why? They optimize for bug fixes in
Python repos—not the auth flows, API integrations, and CRUD dashboards that occupy 80% of real dev work.
So we built VibeCodingBench: 180 tasks across SaaS features, glue code, AI integration, frontend, API integrations, and code evolution.
Multi-dimensional scoring: Functional (40%) + Visual (20%) + Quality (20%) - Cost/Speed penalties. Security gate: Any OWASP Top 10 vuln = automatic 0.
Top 5 Results (Jan 2026):
Claude Opus 4.5 — 89.2% | $12.31 | 44s
Claude Haiku 4.5 — 89.0% | $3.03 | 22s
Grok 4 Fast — 88.8% | $0.21 | 70s
4⃣ OpenAI GPT-5.2 — 88.8% | $5.01 | 28s
5⃣ Qwen3 Max — 88.6% | $5.42 | 45s
The real story? Cost varies 60x between similar performers. Grok 4 Fast matches GPT-5.2 at 1/25th the cost. Claude Haiku 4.5 delivers near-Opus quality
for $3 total.
Live dashboard: https://vibecoding.llmbench.xyz/
GitHub repo: https://github.com/alt-research/vibe-coding-benchmark-public
Thesis: https://github.com/alt-research/vibe-coding-benchmark-public/blob/main/docs/THESIS.md
The ultimate test isn't fixing a bug in scikit-learn. It's shipping a feature your users need—safely, efficiently—before the sprint ends.
Open source. Contributions welcome.
Re:Did Blockchain Become Obsolete? Demystifying Hashgraph: Benefits an...
on Thursday November 09, 2017 @02:03AM (#15660455)
Basically, it is unfair to compare permissioned blockchain to public blockchains.