Yao Dexter
@dextery
Yao Dexter
The art of compaction lies in the selection of what to keep versus what to discard, as overly aggressive compaction can result in the loss of subtle but critical context whose importance only becomes apparent later.
anthropic.com • Effective Context Engineering for AI Agents
As skills grow in complexity, they may contain too much context to fit into a single SKILL.md , or context that’s relevant only in specific scenarios. In these cases, skills can bundle additional files within the skill directory and reference them by name from SKILL.md . These additional linked files are the third level (and beyond) of detail,... See more
Anthropic Engineering • Equipping Agents for the Real World With Agent Skills
One of the most intelligent case studies in design is the Chinese tea cup. They’re made without handles simply because if it’s too hot to touch, it’s too hot to drink.
Humans naturally want to add more. Add a cardboard sleeve, add a warning on the outside of the cup, add a handle. The result of all these things never cools down the actual contents.
... See more
This approach mirrors human cognition: we generally don’t memorize entire corpuses of information, but rather introduce external organization and indexing systems like file systems, inboxes, and bookmarks to retrieve relevant information on demand.
anthropic.com • Effective Context Engineering for AI Agents
Letting agents navigate and retrieve data autonomously also enables progressive disclosure—in other words, allows agents to incrementally discover relevant context through exploration. Each interaction yields context that informs the next decision: file sizes suggest complexity; naming conventions hint at purpose; timestamps can be a proxy for... See more
anthropic.com • Effective Context Engineering for AI Agents
Agents also tend to grapple with natural language names, terms, or identifiers significantly more successfully than they do with cryptic identifiers. We’ve found that merely resolving arbitrary alphanumeric UUIDs to more semantically meaningful and interpretable language (or even a 0-indexed ID scheme) significantly improves Claude’s precision in... See more
anthropic.com • Writing Effective Tools for Agents — With Agents
（在需要有一定泛化能力的场景下）Give goal, workflows, frameworks, or guidelines, rather than hard-coded rules
At startup, the agent pre-loads the name and description of every installed skill into its system prompt.
This metadata is the first level of progressive disclosure : it provides just enough information for Claude to know when each skill should be used without loading all of it into context. The actual body of this file is the second level of... See more
This metadata is the first level of progressive disclosure : it provides just enough information for Claude to know when each skill should be used without loading all of it into context. The actual body of this file is the second level of... See more