skills.rest
rierino-open
rierino-openOFFICIAL
Explore all skills by this author →
WebsiteGitHub

ai_agent_assistant

Generates schema-compliant GenAI agent configuration JSON for the Rierino platform.

1|Updated Mar 27, 2026
Search Tags:#llm#system prompt#agent#configuration#guardrails#genai#rierino
One-click install
npx skills add https://github.com/rierino-open/rierino-claude-plugin --skill ai-agent-assistant-rierino-open
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai_agent_assistant
Source: https://github.com/rierino-open/rierino-claude-plugin/tree/main/development/skills/ai_agent_assistant
Command: npx skills add https://github.com/rierino-open/rierino-claude-plugin --skill ai-agent-assistant-rierino-open

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Creating a production-ready GenAI agent on the Rierino platform requires assembling a complex JSON record with model settings, memory strategy, tool sagas, guardrails, and a specialist system prompt, which is error-prone when done by hand. ## Core Features & Use Cases - Agent Record Generation: Translates natural-language role and behavior requirements into a complete, schema-compliant agent JSON object with model parameters, memory, and tool sagas. - Guardrail Configuration: Always generates proportional guardrails including a risk policy, jailbreak protection, and PII masking matched to the agent's risk level. - System Prompt Authoring: Writes specialist, second-person system prompts with structured sections such as Role & Expertise, Domain Rules, and Validation Checklist. - Use Case: Ask for a sales assistant that can query CRM data, and receive a validated agent record with real saga IDs fetched from the platform catalog, ready to save after your approval. ## Quick Start Ask the assistant to create a Rierino GenAI agent for a specific role, describing its purpose, tools, and model settings.

Frequently Asked Questions about ai_agent_assistant

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a GenAI agent on the Rierino platform?▼

Describe the agent's role, purpose, behavior, and model settings in natural language. The skill generates a complete schema-compliant agent JSON record with instructions, guardrails, memory, and tool sagas, which you approve before it is saved.

How do I attach tool sagas to a Rierino AI agent?▼

The skill fetches the available saga catalog from the Rierino MCP server and selects valid saga IDs for the agent's toolSagas field. Saga IDs are never invented; if none apply, the array is left empty.

What guardrails does a generated Rierino agent include?▼

Every agent includes a risk policy with block thresholds, jailbreak protection via regex-matcher input guardrails, and PII masking via regex-masker. Tool-enabled and sensitive-domain agents receive additional tool-response and output guardrails.

Can I modify an existing Rierino agent instead of creating one?▼

Yes, the skill reads the existing agent record by ID using the Rierino MCP tools, applies your requested changes, and preserves the original agent ID. Changes are only saved after your explicit approval.

Which model providers work with the default Rierino agent configuration?▼

The default OpenAiChatModel class works with any provider offering OpenAI-compatible APIs. For non-OpenAI providers, a baseUrl parameter pointing to the provider's compatible endpoint is added to the methods configuration.

Related Skills

Explore More Software Engineering Skills→
slowman2084slowman2084
6

platform_react_runner

Run React Agent prompt tests on external platforms with LangGraph and MCP tooling.

Community
Advanced
ValescoAgencyValescoAgency

run-attested

Trigger the AFK runner for attested contracts and stream audit output.

Official
Advanced
LauraFlorentinLauraFlorentin
1

guardrails

Inspect inputs and outputs to enforce safety and policy compliance.

Community
Advanced
ModelsLabModelsLab
10

modelslab-chat-generation

Chat with 60+ LLMs through one OpenAI-compatible endpoint.

Official
Advanced