正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Problem: AI coding performance dips when context windows exceed 50% Solution: Combine Cline's context window awareness with the `new_task` tool + .clinerules to create a workflow that autonomously hands off tasks before hitting limits, ensuring persistent memory. A guide: 🧵

Cline

52,457 subscribers

62,196 次观看 • 1 年前 •via X (Twitter)

新闻政治科学技术教育

Anya Rossi• Live Now

Private livecam show

11 条评论

Cline 的头像

Cline1 年前

Large context windows aren't a silver bullet. Models can still struggle or "forget" past ~50% usage, degrading performance. Plus, manually re-explaining project context each time you restart is a major workflow killer.

Cline 的头像

Cline1 年前

The first piece is awareness: Cline is aware of its own context window usage (visible in `environment_details`). It knows how much "memory" is being used relative to the model's limit (e.g., 105k/200k tokens = 53%).

Cline 的头像

Cline1 年前

The second piece is the `new_task` tool. This allows Cline to cleanly end the current session and immediately start a fresh one, crucially preloading it with specific context you define (summaries, next steps, file states, etc.).

Cline 的头像

Cline1 年前

The magic happens when you combine these in `.clinerules`. You define the trigger (e.g., "if context 50%, propose handoff") and exactly what context Cline should package using `new_task`. This creates an automated, proactive context management workflow.

Cline 的头像

Cline1 年前

The outcome? Cline intelligently manages its own context before performance degrades. No more manual resets or tedious re-explaining. For complex, multi-session tasks, it feels like working with an agent that has persistent memory.

Cline 的头像

Cline1 年前

Ready to build workflows that beat context limits? Learn how to implement this with `.clinerules` and the `new_task` tool in our docs:

NICE 的头像

NICE1 年前

Stay competitive by balancing cutting-edge AI with automation tools. Forrester shows how.

Cline 的头像

Cline1 年前

Try Cline today 👇

Jonathan Chang 的头像

Jonathan Chang1 年前

this is very cool. To solve this same issue, I created time travel tool to allow agent to partially clear the conversation and summarize it. I think it allow a more flexible way to manage context and make continuation more seamless.

Dexter 的头像

Dexter1 年前

Just curious: how are you measuring context windows? Simply token count input from the user?

Dan 的头像

Dan1 年前

cline just keeps consistently delivering. no wonder it's my go-to.

相关视频

LLMs need focus, because attention isn't enough. (thread on the Focus Chain, Cline's link to persistent context)

LLMs need focus, because attention isn't enough. (thread on the Focus Chain, Cline's link to persistent context)

Cline

26,570 次观看 • 9 个月前

1/ The Model Context Protocol isn't just another dev tool - it's letting AI assistants break free from chat windows to directly manage your Git repos, run tests, and maintain project memory. Here's why MCP is a game-changer 🧵

1/ The Model Context Protocol isn't just another dev tool - it's letting AI assistants break free from chat windows to directly manage your Git repos, run tests, and maintain project memory. Here's why MCP is a game-changer 🧵

Cline

50,041 次观看 • 1 年前

Question we get constantly: "How does Cline handle context limits in long-running tasks?" Here's how users can manage their context: /newtask: Creates a detailed handoff summary and starts fresh context. Like handing off work to a new engineer with full background. more 👇

Question we get constantly: "How does Cline handle context limits in long-running tasks?" Here's how users can manage their context: /newtask: Creates a detailed handoff summary and starts fresh context. Like handing off work to a new engineer with full background. more 👇

Cline

21,739 次观看 • 10 个月前

Devs are using Memory Bank as a project architecture and planning tool before they write any code. "I prefer starting in Cline, using Memory Bank to build out the context files as a roadmap, and letting Cline take the wheel." How to use Memory Bank for project planning 🧵

Devs are using Memory Bank as a project architecture and planning tool before they write any code. "I prefer starting in Cline, using Memory Bank to build out the context files as a roadmap, and letting Cline take the wheel." How to use Memory Bank for project planning 🧵

Cline

96,628 次观看 • 1 年前

AI coding agents hit a wall when codebases get massive. Even with 2M token context windows, a 10M line codebase needs 100M tokens. The real bottleneck isn't just ingesting code - it's getting models to actually pay attention to all that context effectively.

AI coding agents hit a wall when codebases get massive. Even with 2M token context windows, a 10M line codebase needs 100M tokens. The real bottleneck isn't just ingesting code - it's getting models to actually pay attention to all that context effectively.

Garry Tan

976,161 次观看 • 11 个月前

A beginner's guide to context awareness 🧠

A beginner's guide to context awareness 🧠

Windsurf

34,829 次观看 • 1 年前

built a small CLI tool that moves your AI coding session context between Claude Code, Gemini, Copilot, Codex and OpenCode when you hit rate limits. no more re-explaining what you were doing, or painful copy/paste from terminal that lose tool-call context. just `npx continues`

built a small CLI tool that moves your AI coding session context between Claude Code, Gemini, Copilot, Codex and OpenCode when you hit rate limits. no more re-explaining what you were doing, or painful copy/paste from terminal that lose tool-call context. just `npx continues`

Yigit Konur

213,347 次观看 • 3 个月前

FYI: You can use Cline as an orchestrator to spawn Claude Code subagents that run tasks in parallel. "spawn 3 subagents to investigate cline's repo and have them create a single report on API providers, cline's tools, and how cline uses the browser"

FYI: You can use Cline as an orchestrator to spawn Claude Code subagents that run tasks in parallel. "spawn 3 subagents to investigate cline's repo and have them create a single report on API providers, cline's tools, and how cline uses the browser"

Cline

33,047 次观看 • 11 个月前

Tired of your AI assistant getting amnesia? Our community built something fascinating: 🧠 Memory Bank. It's like giving your AI a perfectly organized brain. Through custom instructions, Cline maintains a living documentation system that rebuilds context after every reset. The secret sauce? .clinerules - a learning journal where Cline captures patterns, preferences, and project intelligence. No more repeating yourself. Watch it remember everything across sessions👇

Tired of your AI assistant getting amnesia? Our community built something fascinating: 🧠 Memory Bank. It's like giving your AI a perfectly organized brain. Through custom instructions, Cline maintains a living documentation system that rebuilds context after every reset. The secret sauce? .clinerules - a learning journal where Cline captures patterns, preferences, and project intelligence. No more repeating yourself. Watch it remember everything across sessions👇

Cline

36,232 次观看 • 1 年前

Developers: MCP is about to become extremely relevant as AI needs to connect with real-world data. The problem? Most don't know how to build MCP plugins yet. Here's how to build them with Cline in 3 simple steps using our .clinerules protocol. Thread 🧵

Developers: MCP is about to become extremely relevant as AI needs to connect with real-world data. The problem? Most don't know how to build MCP plugins yet. Here's how to build them with Cline in 3 simple steps using our .clinerules protocol. Thread 🧵

Cline

189,031 次观看 • 1 年前

Introducing Inspector's context-aware screenshot tool! Screenshots as context for your coding agent is a pain. I decided I was tired of it and shipped something today to help with that.

Introducing Inspector's context-aware screenshot tool! Screenshots as context for your coding agent is a pain. I decided I was tired of it and shipped something today to help with that.

mikeBuildsMore

21,932 次观看 • 7 个月前

New short course: LLMs as Operating Systems: Agent Memory, created with Letta, and taught by its founders Charles Packer and Sarah Wooders. An LLM's input context window has limited space. Using a longer input context also costs more and results in slower processing. So, managing what's stored in this context window is important. In the innovative paper MemGPT: Towards LLMs as Operating Systems, its authors (which include the instructors) proposed using an LLM agent to manage this context window. Their system uses a large persistent memory that stores everything that could be included in the input context, and an agent decides what is actually included. Take the example of building a chatbot that needs to remember what's been said earlier in a conversation (perhaps over many days of interaction with a user). As the conversation's length grows, the memory management agent will move information from the input context to a persistent searchable database; summarize information to keep relevant facts in the input context; and restore relevant conversation elements from further back in time. This allows a chatbot to keep what's currently most relevant in its input context memory to generate the next response. When I read the original MemGPT paper, I thought it was an innovative technique for handling memory for LLMs. The open-source Letta framework, which we'll use in this course, makes MemGPT easy to implement. It adds memory to your LLM agents and gives them transparent long-term memory. In detail, you’ll learn: - How to build an agent that can edit its own limited input context memory, using tools and multi-step reasoning - What is a memory hierarchy (an idea from computer operating systems, which use a cache to speed up memory access), and how these ideas apply to managing the LLM input context (where the input context window is a "cache" storing the most relevant information; and an agent decides what to move in and out of this to/from a larger persistent storage system) - How to implement multi-agent collaboration by letting different agents share blocks of memory This course will give you a sophisticated understanding of memory management for LLMs, which is important for chatbots having long conversations, and for complex agentic workflows. Please sign up here!

New short course: LLMs as Operating Systems: Agent Memory, created with Letta, and taught by its founders Charles Packer and Sarah Wooders. An LLM's input context window has limited space. Using a longer input context also costs more and results in slower processing. So, managing what's stored in this context window is important. In the innovative paper MemGPT: Towards LLMs as Operating Systems, its authors (which include the instructors) proposed using an LLM agent to manage this context window. Their system uses a large persistent memory that stores everything that could be included in the input context, and an agent decides what is actually included. Take the example of building a chatbot that needs to remember what's been said earlier in a conversation (perhaps over many days of interaction with a user). As the conversation's length grows, the memory management agent will move information from the input context to a persistent searchable database; summarize information to keep relevant facts in the input context; and restore relevant conversation elements from further back in time. This allows a chatbot to keep what's currently most relevant in its input context memory to generate the next response. When I read the original MemGPT paper, I thought it was an innovative technique for handling memory for LLMs. The open-source Letta framework, which we'll use in this course, makes MemGPT easy to implement. It adds memory to your LLM agents and gives them transparent long-term memory. In detail, you’ll learn: - How to build an agent that can edit its own limited input context memory, using tools and multi-step reasoning - What is a memory hierarchy (an idea from computer operating systems, which use a cache to speed up memory access), and how these ideas apply to managing the LLM input context (where the input context window is a "cache" storing the most relevant information; and an agent decides what to move in and out of this to/from a larger persistent storage system) - How to implement multi-agent collaboration by letting different agents share blocks of memory This course will give you a sophisticated understanding of memory management for LLMs, which is important for chatbots having long conversations, and for complex agentic workflows. Please sign up here!

Andrew Ng

200,673 次观看 • 1 年前

New in Cline 3.7.0: The .clinerules folder -- a better way to organize your project rules. Here's how you can implement a modular rules system that transforms how Cline understands your projects: 🧵

New in Cline 3.7.0: The .clinerules folder -- a better way to organize your project rules. Here's how you can implement a modular rules system that transforms how Cline understands your projects: 🧵

Cline

67,191 次观看 • 1 年前

all AI coding agents hallucinate when it comes to APIs here's a solution for this: Context Hub by Andrew Ng, completely open-source i tested it myself, watch the video to see the live demo

all AI coding agents hallucinate when it comes to APIs here's a solution for this: Context Hub by Andrew Ng, completely open-source i tested it myself, watch the video to see the live demo

Nidhi Singh

32,801 次观看 • 2 个月前

Introducing Ambient Context Aware AI on your computer. I have just scraped the whole Discord with it and made a summary of pain points ppl have Features: - Sees the whole screen ( even parts you need to scroll to) - long context window - Has memory Link:

Introducing Ambient Context Aware AI on your computer. I have just scraped the whole Discord with it and made a summary of pain points ppl have Features: - Sees the whole screen ( even parts you need to scroll to) - long context window - Has memory Link:

Robert Lukoszko

15,383 次观看 • 1 年前

In this episode, Beyang and Thorsten discuss strategies for effective agentic coding, including the 101 of how it's different from coding with chat LLMs, the key constraint of the context window, how and where subagents can help, and the new oracle subagent which combines multiple LLMs. 00:38 Intros 03:20 How coding with agents is very different from coding with prior AI tools 10:31 Example: fix a simple issue 14:13 Example: debugging an issue with an MCP server 21:50 Example: unifying two build scripts 25:09 How the context window is a key constraint 31:01 Why it's best to focus on one thing at a time 33:09 Subagents and context windows 33:49 The codebase search subagent 38:33 General-purpose subagents 44:05 When to use subagents 46:44 The oracle subagent and o3 51:32 Multi-model agents

In this episode, Beyang and Thorsten discuss strategies for effective agentic coding, including the 101 of how it's different from coding with chat LLMs, the key constraint of the context window, how and where subagents can help, and the new oracle subagent which combines multiple LLMs. 00:38 Intros 03:20 How coding with agents is very different from coding with prior AI tools 10:31 Example: fix a simple issue 14:13 Example: debugging an issue with an MCP server 21:50 Example: unifying two build scripts 25:09 How the context window is a key constraint 31:01 Why it's best to focus on one thing at a time 33:09 Subagents and context windows 33:49 The codebase search subagent 38:33 General-purpose subagents 44:05 When to use subagents 46:44 The oracle subagent and o3 51:32 Multi-model agents

Amp — Research Preview

24,534 次观看 • 10 个月前

🚨 KEEP CALM AND AI - AUTONOMOUS AGENTS WITH INFINITE MEMORY We are launching the ability to create arbitrary agents that run on schedule and have access to a persistent and infinite memory The agents will be able to store, retrieve, and update information across sessions and perform repetitive tasks that require persistent memory Another step towards AGI and automating white-collar work 🚀🚀🚀

🚨 KEEP CALM AND AI - AUTONOMOUS AGENTS WITH INFINITE MEMORY We are launching the ability to create arbitrary agents that run on schedule and have access to a persistent and infinite memory The agents will be able to store, retrieve, and update information across sessions and perform repetitive tasks that require persistent memory Another step towards AGI and automating white-collar work 🚀🚀🚀

Bindu Reddy

12,626 次观看 • 4 个月前

Mermaid diagrams are the hidden superpower in AI-human communication. They're intuitive for both LLMs to generate and humans to understand, creating a shared visual language that bridges the coding gap. Why does this matter? When working with LLMs, there's usually a translation problem: >You explain what you want in natural language >The AI internally plans complex workflows >You're left wondering "what exactly is it doing?" What makes Cline's implementation so powerful is that it visualizes its planning mode through Mermaid diagrams. This transparency means you can actually SEE how Cline will approach your implementation before it starts. No more black box. No more guessing. Cline thinks through complex processes using Mermaid, showing you a visual map of what it's building. This creates shared understanding that eliminates misalignment. Want to make AI more useful? Give it a visual language both sides understand.

Mermaid diagrams are the hidden superpower in AI-human communication. They're intuitive for both LLMs to generate and humans to understand, creating a shared visual language that bridges the coding gap. Why does this matter? When working with LLMs, there's usually a translation problem: >You explain what you want in natural language >The AI internally plans complex workflows >You're left wondering "what exactly is it doing?" What makes Cline's implementation so powerful is that it visualizes its planning mode through Mermaid diagrams. This transparency means you can actually SEE how Cline will approach your implementation before it starts. No more black box. No more guessing. Cline thinks through complex processes using Mermaid, showing you a visual map of what it's building. This creates shared understanding that eliminates misalignment. Want to make AI more useful? Give it a visual language both sides understand.

Cline

36,786 次观看 • 1 年前

⚒️ How to build an AI agent with persistent memory? Watch Shaw (spirit/acc) building an agent from scratch with elizaOS & showing how external, queryable memory based on OriginTrail gives agents context and extends their memory beyond a single session. 02:15 – Set up your first agent 08:55 – Initializing runtime & memory foundations 12:50 – Wiring message handling & agent memory 24:45 – Persistent conversational memory in action 38:00 – Extending agents with plugins

⚒️ How to build an AI agent with persistent memory? Watch Shaw (spirit/acc) building an agent from scratch with elizaOS & showing how external, queryable memory based on OriginTrail gives agents context and extends their memory beyond a single session. 02:15 – Set up your first agent 08:55 – Initializing runtime & memory foundations 12:50 – Wiring message handling & agent memory 24:45 – Persistent conversational memory in action 38:00 – Extending agents with plugins

OriginTrail

37,143 次观看 • 4 个月前

Sam Altman says the next big thrust is making models handle much longer tasks We have already moved from models handling 5-second coding tasks with GPT-3.5 to 5-hour tasks with GPT-5 The goal now is to integrate enterprise context so AI can handle tasks that require months or years

Sam Altman says the next big thrust is making models handle much longer tasks We have already moved from models handling 5-second coding tasks with GPT-3.5 to 5-hour tasks with GPT-5 The goal now is to integrate enterprise context so AI can handle tasks that require months or years

Haider.

122,717 次观看 • 6 个月前