Tag
作者开源了 agent-engineer 插件，适用于 Claude Code、Codex CLI、Cursor 等，包含 11 个技能、4 个子代理和 5 个钩子；盲测 A/B 显示 Claude Code (Opus 5.5) 使用插件后任务需求满足率从 34% 提升到 53%。
A practitioner asks how agent frameworks should handle write-tool timeouts, weighing idempotency keys, read-back verification, and human escalation when downstream APIs lack idempotency support.
LangChain is hosting a workshop on agent engineering for life sciences, focusing on creating accurate, observable, and reliable research workflows.
The article explores the challenges of delegation and permission inheritance in agentic systems, raising questions about autonomy and authority management when agents spawn sub-agents.
LangChain is hosting an evening event in San Francisco featuring technical talks on continual learning, speakers from Prime Intellect and Baseten, and networking opportunities for the AI agent engineering community.
LangChain announces a free Academy course on building AI agents with Python, covering foundational concepts like react loops, MCP servers, and human-in-the-loop.
Cursor's engineering blog describes how they built a development environment for cloud agents, introducing the anydev CLI and the Cloud Doctor self-healing mechanism, enabling agents to now contribute more than half of merged PRs.
Unused tools in an AI agent's toolset still consume tokens and add noise to tool selection, so agents should load only the tools required for the current task.
Chinese researchers achieved a breakthrough in autonomous AI agent engineering with a code-as-harness paradigm that replaces text prompts with executable verification substrates, enabling deterministic multi-agent execution through six internal processes.
LangChain Academy launches the first certification for the full Agent Development Lifecycle, offering a 50% discount for the first two months.
The article explains the 'Tool Rot Paradox' where installing many static agent skills causes context window degradation and security issues in production, and advocates for a dynamic discovery approach using a meta-skill that fetches tools on demand to keep the system prompt lean.
The author reflects on how long-running AI agents encounter failures unrelated to the initial prompt, arguing that environment design (tools, docs, validation, architecture rules) matters more. They discuss concepts like harness engineering, keeping AGENTS.md small, using linters, and evaluator agents, while noting the cost trade-offs.
This article argues that building AI agents is easy, but providing them with sufficient business context remains the hard and unsolved infrastructure problem, proposing the need for a dedicated 'context layer' akin to GitHub for code.
An engineer tracked his coding agent's token usage over a week, finding that only 0.67% of tokens were spent on actual tasks, with 99% consumed by tool directories, skill descriptions, and system prompts. He provides optimization strategies, including shell output filtering which saved 46.9% of tokens.
This article explores the trend of upgrading prompt engineering to Agent engineering, emphasizing structured context management of AI agents through methods like AGENTS.md, and shares a minimal closed-loop workflow methodology.
Recommending Harrison Chase's blog post on multi-agent system design. The key insight is that multi-agent systems split into 'reading' and 'writing' tasks; reading is easy to parallelize while writing leads to conflicts. It also summarizes suitable and unsuitable scenarios and engineering prerequisites.
Former OpenAI engineer Ryan Lopopolo joins Google Cloud as Chief Agent Engineer, bringing OpenAI's Agent engineering methodology to the cloud platform.
Microsoft open-sourced SkillOpt, a framework that treats agent skill documents as trainable artifacts, optimizing them through iterative feedback loops without modifying model weights. It is available as a Python library with support for self-evolving agent skills.
作者分享了在GIAC 2026深圳会议上关于Agent Engineering的演讲，强调构建可靠、有状态的代理的重要性。
A thread explaining the 5 core mental shifts needed to transition from traditional software engineering to agent engineering, emphasizing why conventional patterns like hard-coded routes and binary tests fail with AI agents.