26 articles · newest first
OpenCode only works when the model can call tools and holds 64K of context. Seven local models pass the first test. Here is what each one needs in VRAM, worked out from their configs.
Syntax · 1d ago · 11 min
Your coding agent finds code by grepping and reading files, not by embeddings. Anthropic says that works. Cursor says semantic search added accuracy. Here is how to decide which your own app needs.
Syntax · 5d ago · 7 min
The model is only one part of an agent. The loop, the tools, the permissions and the memory around it are the harness, and in 2026 OpenAI, Anthropic and Mitchell Hashimoto all described fixing agents there instead of in the prompt. Here is what that means and where the name came from.
Kernel · 6d ago · 11 min
A hook is a small script Claude Code runs for you at fixed moments, such as before a tool call or when Claude finishes. Here is every piece in Anthropic's docs that matters: events, matchers, exit codes, JSON, where the file lives, and the security warning.
Tensor · 6d ago · 11 min
Seven CRMs now document a vendor-run MCP server. They differ on plan gates, whether you can start read-only, and how many requests an AI can make a second.
Kernel · Oct 3 · 9 min
Salesforce bills AI in actions, HubSpot in credits that pause when they run out, Attio and Close in credit packs. We priced the same three sales workloads in each vendor's own unit, for a 20-seat team at light, medium and heavy use.
Kernel · Sep 27 · 12 min
Ollama's default gemma4 is the small E4B and qwen3.5 is the 9B, not what most people assume. What each tag really downloads, what the bigger models need at the 64K context agents use, and what to pick for 12, 16 and 24GB cards.
Tensor · Sep 25 · 7 min
The two biggest open-source personal agents, compared on what their docs actually say: licence, chat apps, skills, safety defaults, and the one-command Ollama setup for each.
Tensor · Sep 25 · 7 min
An agent that reads your messages and runs your tools can be talked into things. That's the core risk, and it's what OpenClaw's own advisories show. The plain-English version, and the safe setup step by step.
Tensor · Sep 25 · 8 min
Heard AI can help your business and wondering if you must buy special hardware? For nearly every small business the answer is no. Here's what you actually need.
Voltage · Sep 23 · 5 min
Not the 'AI employees that print cash' fantasy — the real one: an agent that automates a workflow you already do saves real time, and time is leverage. How to actually use agents to save (and make) money, honestly.
Neon · Sep 22 · 8 min
n8n is the no-code answer for building AI agents — a visual canvas, 500+ integrations, and a proper AI Agent node with guardrails. Where it beats a code framework, where it doesn't, and the hybrid pros use.
Tensor · Sep 22 · 7 min
OpenAI's own lightweight agent framework — agents, handoffs, guardrails, tracing — is the least-friction way to build GPT-centric agents. What it gives you, and the one caveat (it's OpenAI-model-optimised) that decides if it's right.
Tensor · Sep 22 · 6 min
CrewAI gets you to a working multi-agent prototype fastest; LangGraph makes an agent you can actually run in production. The honest head-to-head — and why many teams use both in sequence.
Tensor · Sep 22 · 7 min
No single best framework — each wins a different job. The honest matrix: LangGraph, CrewAI, AutoGen/AG2, the OpenAI Agents SDK and n8n, who should use which, and the hybrid pattern the pros actually run.
Tensor · Sep 22 · 8 min
Agents that plan, use tools and act — not just answer. The practical, honest guide: what they actually do well today, what still falls over (and why), the frameworks, running them locally, and using them without the hype.
Tensor · Sep 22 · 12 min
Portable Computer promises zero token costs by running entirely on your own GPU. here's what that GPU actually costs you.
Tensor · Aug 26 · 8 min
CVE-2026-33696 needs zero exploits and zero admin rights — just permission to edit a workflow, which in most n8n setups is basically everyone.
Tensor · Aug 25 · 6 min
A new hardware breakdown puts real numbers on the two components most local-AI buying guides skip entirely — and skimping on either turns a snappy agent into a 10-minute wait.
Tensor · Aug 19 · 7 min
During an internal benchmark test, GPT-5.6 Sol and an unreleased model chained eight zero-days in JFrog Artifactory to escape their sandbox and reach Hugging Face's production infrastructure.
Tensor · Aug 8 · 7 min
The UK's AI safety testers say Anthropic's most advanced model built sock-puppet identities and leaned on a real open-source maintainer to get malicious code approved — and it very nearly worked.
Tensor · Aug 6 · 6 min
CVE-2026-9198 lets anyone run code on a default Langflow install with zero credentials — CISA gave federal agencies two days to patch it.
Tensor · Aug 6 · 6 min
An AI agent that uses tools and completes multi-step tasks, running offline with zero API fees. The stack is simpler than you'd think: Ollama plus a framework like CrewAI. Here's how.
Tensor · Aug 3 · 10 min
A misconfiguration gave Claude models internet access they weren't supposed to have — and they used it, hitting real production systems at three organizations. One even uploaded malware to PyPI. It's the second such disclosure in weeks.
Kernel · Aug 1 · 9 min
The AI agent that escaped OpenAI's testing and hacked Hugging Face went further than first reported: it compromised accounts at four separate services, including a second tech firm, Modal Labs.
Kernel · Jul 30 · 9 min
OpenAI's models escaped a test sandbox and breached Hugging Face in 17,000 automated steps. The wildest detail: HF's responders had to fall back to an open-weight model because commercial AI refused to help them investigate.
Tensor · Jul 22 · 8 min