Lesson 936 of 1596
Tool Discovery Strategies for Long-Lived Agents
How to give an agent access to 200+ tools without blowing the context window.
Creators · Agentic AI · ~7 min read
The premise
Loading every tool into the prompt is wasteful — retrieve only the tools the current task needs.
What AI does well here
- Embed tool descriptions and retrieve top-k by query.
- Group tools into bundles selectable by task type.
- Cache the chosen toolset across multi-step tasks.
What AI cannot do
- Guarantee the right tool is in the top-k for novel queries.
- Replace a human-curated allowlist for high-risk tools.
Key terms in this lesson
Practice this safely
Use a small project example from your own work. The useful move is to compare the AI's draft against your goal, sources, and constraints before you trust it.
- 1Ask AI to explain tool discovery in plain language, then underline anything that sounds uncertain or too broad.
- 2Give it one detail from "Tool Discovery Strategies for Long-Lived Agents" and ask for two possible next steps plus one reason each step might be wrong.
- 3Check retrieval over tools against a trusted source, teacher, adult, expert, or original document before you use it.
End-of-lesson quiz
Check what stuck
10 questions · Score saves to your progress.
Tutor
Curious about “Tool Discovery Strategies for Long-Lived Agents”?
Ask anything about this lesson. I’ll answer using just what you’re reading — short, friendly, grounded.
Progress saved locally in this browser. Sign in to sync across devices.
Related lessons
Keep going
Creators · 48 min
Computer Use API: Letting AI Click Through GUIs
Computer Use lets Claude see your screen and use it — mouse, keyboard, apps. The capability is real, the gotchas are real. A hands-on look at what works in 2026.
Creators · 45 min
Browser Agents: Capabilities and Pitfalls
Browser agents — Operator, Atlas, Browser Use, MultiOn — are the most visible agent category. The capability is genuine, the failure modes are specific. Build with eyes open.
Creators · 75 min
Capstone: Build and Ship a Real Agent
Everything comes together. Design, code, test, secure, and ship a production-quality agent with open-source code you can fork today.