Manage agent instances and ZFAE weight banks in a0-betatest.
npx skills add https://github.com/The-Interdependency/skill-lib --skill a0p-instancing
Please help me install this Agent Skill. Skill: a0p-instancing Source: https://github.com/The-Interdependency/skill-lib/tree/main/a0p-instancing Command: npx skills add https://github.com/The-Interdependency/skill-lib --skill a0p-instancing
This Skill provides a methodology for managing agents in the a0-betatest environment, addressing the need for CRUD operations on AgentInstances and CharacterSheets, and handling per-instance native ZFAE weight banks and their training/distillation loops.
Use the a0p-instancing skill to create a new AgentInstance with a CharacterSheet and initialize its native ZFAE weight bank.
High-intent search queries and answers about installing and using this skill.
You can manage agent instances and character sheets in a0-betatest by performing CRUD operations to create, update, and delete agents alongside their associated configuration profiles.
You can train and persist native ZFAE weight banks by initializing a weight bank for an agent instance and utilizing checkpointing mechanisms to ensure data integrity during the training and distillation loops.
Available agent inference modes in the a0-betatest environment include ZFAE_NATIVE and ZFAE_ASSISTED, allowing you to select different processing requirements for your agent operations.
Yes, you need the a0-betatest backend and related tools to execute agent CRUD operations, initialize native ZFAE weight banks, and manage checkpointing and persistence.
Sentinel evaluations and pending-override gates provide operational governance by intercepting and validating agent actions, ensuring that inference modes and updates adhere to defined safety constraints.
Agent state management requires checkpointing during distillation loops to persist training progress and prevent data loss, ensuring native ZFAE weight banks maintain integrity across operational cycles.
trueraCreate and curate ground-truth evaluation datasets for TruLens.
jluo41Package trained AI models into versioned assets with training results.
jackspaceOptimize high-throughput PPO+LSTM reinforcement learning training with vectorized environments.
QcasaresManage entities, relationships, and constraints in a typed knowledge graph.
