Securing Claude Agents: Sandboxing, Secrets, Injection Defense
Harden Claude agents: sandboxing, least privilege, secrets handling, and prompt-injection defense for tool-using systems.
7 min read
Agentic AI & LLMs
Agentic AI & LLMs archive page 558 of 631
Agentic AI & LLMs
Agentic AI, LLM engineering, and the models behind modern automation — multi-agent systems, LLM evaluation and comparisons, RAG, fine-tuning, AI infrastructure, security, and production AI engineering.
From the blog
Latest articles
Cutting Claude Agent Cost: Caching, Batching, Fast Runs
Token-cost engineering for Claude agents: prompt caching, batching, model routing, and context discipline to keep runs cheap and fast.
7 min read
Debugging Claude Agents: Loops, Bad Tool Calls, Hallucinated Args (Cowork Plugins Across Enterprise)
Debug Claude agents and Cowork plugins: catch infinite loops, wrong tool calls, and hallucinated arguments before they wreck a run.
8 min read
Prompt and Context Design for Cowork Plugins
What to put in a Claude Cowork plugin's context and what to leave out: the three context layers, a decision tree, and post-launch tuning.
6 min read
Wiring MCP Servers Into Claude Cowork Plugins
Wire MCP servers into Claude Cowork plugins: scoped auth, strict schemas, legible error handling, and idempotent writes for enterprise tools.
6 min read
Reusable Patterns for Claude Cowork Plugin Design
Code-level patterns for Claude Cowork plugins: narrow skills, progressive disclosure, typed tools, context hygiene, and idempotent writes.
6 min read
Build a Claude Cowork Plugin: Step-by-Step Guide
Build a Claude Cowork plugin from scratch: scope it, lay out the bundle, write the skill, wire an MCP connector, add a sub-agent, and publish.
7 min read
Inside Claude Cowork Plugins: Enterprise Architecture
How Claude Cowork plugins work end to end: skills, MCP connectors, sub-agents, routing, and context isolation across enterprise teams.
8 min read
Showcasing LLM Performance: How Research Papers Present Evaluation Results
Showcasing LLM Performance: How Research Papers Present Evaluation Results
2 min read