메뉴
SGR-enhanced pool of cheap specialized subagents based on sgr-agent-core. Uses Schema-Guided Reasoning to boost mini/Flash models to near-Sonnet quality. Trigger: "sgr", "sgr pool", "cheap agent", "mini agent", "sgr-researcher", "sgr-verifier", "sgr-summarizer".
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Automated research subagent. Accepts a natural-language query, gathers sources from the knowledge graph (KG + BM25 + E5 + Web + Notion), performs SGR synthesis, generates validated knowledge cards, and writes them to the graph. Explicit trigger: "autosearch" — activates the subagent without confirmation. Soft triggers (require confirmation): "research", "analyze", "compare", "deep dive", "review".
SGR-guided knowledge graph node creation with canonical schema validation. Use when creating new nodes, updating existing nodes, or auditing graph health. Enforces closed vocabularies for node types, clusters, and relation types to prevent taxonomy drift. Triggers on: "create node", "add to graph", "new insight", "graph audit", "validate graph", "taxonomy check".
Safe OpenClaw gateway configuration changes with validation and auto-rollback
Design-before-code protocol. Routes new features, refactors, and complex changes through the appropriate design artifact BEFORE implementation begins. Use when the user asks to design a feature, write an RFC, think through architecture, plan a new skill, structure a large refactor, or define the implementation approach before coding. Triggers: "design", "RFC", "architecture", "new skill", "new feature", "large change", "need a plan", "design-architect", "tech spec", "ADR".
Subagent for generating PPTX presentations via Claude Code CLI + official pptx skill. Trigger: "presentation", "slides", "deck".
Multi-subagent framework. Allows the main agent to spawn specialized Claude Code subagents with specific skills. Each subagent runs autonomously (full agentic loop) and returns a structured result. Engine: Claude Code CLI (claude --bare -p). Trigger: "subagent", "spawn agent".
| name | sgr_pool |
| description | SGR-enhanced pool of cheap specialized subagents based on sgr-agent-core. Uses Schema-Guided Reasoning to boost mini/Flash models to near-Sonnet quality. Trigger: "sgr", "sgr pool", "cheap agent", "mini agent", "sgr-researcher", "sgr-verifier", "sgr-summarizer". |
A pool of specialized subagents that use sgr-agent-core as a Python library
for executing research and analytical subtasks with cheap LLMs (gpt-4.1-mini, gemini-2.5-flash, claude-haiku-4-5)
at quality approaching expensive models, thanks to Schema-Guided Reasoning.
| Agent | Task | Default model |
|---|---|---|
sgr-researcher | Search, information gathering, answering questions | gpt-4.1-mini |
sgr-verifier | Fact-checking, claim verification | gpt-4.1-mini |
sgr-summarizer | Summarization, compression | gemini-2.5-flash |
from skills.sgr_pool.executor import SGRPoolExecutor
executor = SGRPoolExecutor()
# Simple call
result = await executor.run("sgr-researcher", "What are the main approaches to knowledge graph construction?")
print(result.answer)
print(result.sources)
print(result.reasoning_steps)
# With a custom model
result = await executor.run("sgr-verifier", "RAG always improves LLM accuracy", model="claude-haiku-4-5")
# With KG access
result = await executor.run("sgr-researcher", "What do we know about knowledge graph architecture?", use_kg=True)
An SGR agent is selected by the orchestrator when:
research | verify | summarizemediumFor complex multi-step research → use AutoResearch (API-first). For code generation → use CLI-first subagents.
02_distill/ (custom)Agent configs: agents/*.yaml
Each yaml defines: model, tools, system prompt, execution limits.
sgr-agent-core>=0.7.0 (pip)TAVILY_API_KEY env var (optional, for web search)Пул спеціалізованих субагентів на базі sgr-agent-core, що використовують Schema-Guided Reasoning для підсилення дешевих LLM (mini/Flash) до якості, наближеної до Sonnet. Три типи агентів: researcher, verifier, summarizer.