Context Window Engineer

$2.99Official

Optimize how context is assembled for LLM calls: system prompt design, few-shot example selection, context pruning, and token budget management.

agent-infrastructurecontext-engineeringprompt-designtoken-managementfew-shotcontext-pruningllm· v1· by SkillingMain
64
Usefulness score
926
Installs
No ratings yet
today
Last updated

Model requirements

Capability tier

advanced

Min context window

33k tokens

Recommended models
Claude Sonnet 4GPT-4o-miniGemini 2.5 FlashLlama 3.3 70B (self-hosted)Qwen 2.5 72B (self-hosted)Mistral Large

Preview

When to use

Use this skill when the user needs to improve an agent's performance by optimizing what goes into its context window: "the agent forgets instructions," "it's not following the format," "context is too long and expensive," "which examples should I include?" Trigger phrases: context engineering, system prompt, few-shot, context window, token budget, prompt assembly, context pruning.

Do NOT use it for building RAG (that's retrieval), agent memory (that's cross-session recall), or for tuning a single prompt's wording in isolation (that's prompt tuning). Context engineering is about the architecture of what fills the window for each call: the structure, selection, ordering, and

🔒 Buy this skill once ($2.99) to unlock the full playbook and install it anywhere — or get everything with Pro.