Tag
11 posts
Palantir AIP is not mainly about connecting an LLM to enterprise data. It is about letting AI agents work inside a governed business world with context, permissions, auditability, and action boundaries.
AI coding agents default to the shortest path: skipping specs, tests, security review, and other practices that make software reliable. Addy Osmani's Agent Skills packages production engineering workflows, quality gates, and best practices into 24 structured skills that agents can consistently follow from idea to launch.
OpenAI previewed the GPT-5.6 family: Sol is the frontier tier, Terra is the balanced everyday tier, and Luna is the fast low-cost tier. The real shift is that model selection moves from 'which model is strongest' to 'which tier should this task be routed to'.
A2A, short for Agent2Agent, is an open protocol for agent interoperability. It focuses on how agents built by different teams, vendors, and runtimes can discover each other, exchange messages, delegate tasks, and collaborate.
MCP, short for Model Context Protocol, is an open protocol for connecting AI applications to external systems. It is not about making the model itself smarter. It standardizes how agents access tools, data sources, prompt templates, and business systems.
Claude Sonnet 5 is Anthropic's next-generation Sonnet model, with stronger reasoning, tool use, coding, and long-horizon task execution. It narrows the gap with Opus 4.8 in agentic scenarios while covering more everyday development and knowledge-work tasks at a lower price.
As the boundaries between engineering, product, design, and data science become thinner, teams may stop organizing work around job titles and start organizing around the kind of value needed at each stage: prototypers, builders, sweepers, growers, and maintainers.
When extending Claude, Skills and Agents are easy to confuse. Their fundamental difference is context: a Skill loads instructions into your current conversation, while an Agent works independently in an isolated context and returns only the result.
Connecting Claude to MCP does not mean it already understands your project. It must call tools first to obtain context. This article explains six prompting techniques for working efficiently with MCP-enabled Claude: establish context, name tools, narrow scope, chain tool calls, use templates, and describe goals.
