Define hierarchical schemas for agent skills, rules, and feedback data.
npx skills add https://github.com/fblissjr/freudagent --skill freud-schema
Please help me install this Agent Skill. Skill: freud-schema Source: https://github.com/fblissjr/freudagent/tree/main/skill Command: npx skills add https://github.com/fblissjr/freudagent --skill freud-schema
Facilitates declarative agent orchestration by providing schema, archetypes, and context assembly for structured task management.
Load the freud-schema skill and use its assemble_runner_context() function to generate contextual prompts for agent tasks.
High-intent search queries and answers about installing and using this skill.
Agent context for hierarchical orchestration is structured by defining schemas, archetypes, and progressive disclosure rules that compose prompts with layered skills and sources. This approach manages versions and domains for structured task execution.
A feedback loop for AI agent skill refinement is implemented using integrated data structures that capture human review and corrections. These structures enable iterative improvements by feeding validation data back into the agent's skill management system.
Runner context for automated extraction workflows is assembled by loading the schema and invoking the assemble_runner_context function. This generates contextual prompts by hierarchically disclosing rules, skills, and sources required for the agent task.
Managing multiple agent skill versions and domains without external dependencies is supported through a declarative schema layer. The system handles version tracking and domain-specific instructions internally within its own data structures.
Progressive disclosure of rules in agent prompt assembly is needed when managing complex, multi-layered tasks that require hierarchical context. It prevents prompt overload by progressively revealing relevant skills and sources based on the orchestration state.
Declarative agent orchestration works for iterative data validation by providing schemas and feedback loops that support automated extraction workflows. This enables continuous correction and skill refinement during AI-powered data processing.
thieungCreate and validate knowledge base vault directories and CLAUDE.md schema.
abhishekdutta18Iteratively refine code search parameters to assemble relevant context for multi-agent workflows.
yanochka11Build browser-based annotation interfaces for reviewing LLM traces with pass/fail labels.
ExpeditoJInject contextual rules into AI prompts across eight processing layers.
