Assemble tiered L0, L1, and L2 contexts into a token-counted bundle.
npx skills add https://github.com/kinnerkarmanish/mak --skill assemble-kinnerkarmanish
Please help me install this Agent Skill. Skill: assemble Source: https://github.com/kinnerkarmanish/mak/tree/main/operations/cognitive-os/skills/assemble Command: npx skills add https://github.com/kinnerkarmanish/mak --skill assemble-kinnerkarmanish
Build a tiered context bundle that gathers and ranks relevant contexts (L0, L1, L2) for a topic, then compacts them into a usable prompt within the AI's context window.
Run the assemble command with a topic to generate a tiered context bundle.
High-intent search queries and answers about installing and using this skill.
To build a tiered context bundle, you locate relevant L0, L1, and L2 contexts for a topic and count their tokens. This compacts multiple context layers into a coherent, ready-to-use prompt that respects the AI's context window.
The best way to load multiple context layers is to assemble them into a tiered bundle at startup or before a call to load background context. This ensures comprehensive grounding by discovering and ranking L0, L1, and L2 contexts with their associated token counts.
Assembling context layers ensures your prompt fits the token count limit by explicitly tracking tokens during the discovery of L0, L1, and L2 contexts. The bundle is then compacted and optimized for the current context window and task.
Yes, you need to integrate L0, L1, and L2 contexts to ground your AI workflow at BUILD mode startup. Assembling a tiered context bundle at this stage seeds project context and ensures comprehensive grounding before task execution.
You should use a tiered context bundle before answering questions about a topic, before a call to load background context, or at BUILD mode startup. This ensures the AI receives a coherent, optimized bundle of relevant information.
Yes, you can use topic context assembly to compact background information into a usable prompt. It gathers and ranks relevant contexts across multiple layers and compacts them into a coherent bundle optimized for the current context window.
zephyrwang6Analyze weekly L1 memory records to extract recurring patterns into L2 behaviors.
lluppesmsPrompt an AI assistant to list required files and their relevance for answering a question.
Svenja-devEstimate token consumption and recommend chunked execution plans for large tasks.
hung-phanOptimize prompts for LLM applications with in-context learning and attack defenses.
