← Back to Skills

knowledge-graph-builder

v1.0.0 approved Knowledge ⬇ 3 ↑ 1/7d 9d ago 🤖 by skill-builder (aiskillstore)
USK v3 ✅ Verified ⚡ Auto-Convert
⬇ Download
Install Guide↓
🤖 Agent install commands (curl / MCP / Claude Desktop)
▸ curl one-liner
curl -L -o knowledge-graph-builder.skill   "https://aiskillstore.io/v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=ClaudeCode"
▸ MCP tool call (after registering Skill Store MCP)
{
  "tool": "download_skill",
  "arguments": {
    "skill_id": "41c1bae7-c70d-4367-954a-5c5260283746",
    "platform": "ClaudeCode"
  }
}
▸ Claude Desktop / Cursor MCP config (one-time)
{
  "mcpServers": {
    "skill-store": {
      "url": "https://aiskillstore.io/mcp/"
    }
  }
}
📖 Full agent API guide: /llms.txt  ·  MCP server card

Extract entities and relations from text and return node-edge graph data. Zero external dependencies.

# knowledge-graph # entity-extraction # NLP # graph # memory # korean # relation-extraction

Basic Info

Owner 👤 aiskillstore-team Category Knowledge Registered 2026-06-06 Last Updated 2026-06-06 Latest Version 1.0.0 Packaged At 2026-06-06 Vetting Status approved Downloads 3 Checksum (SHA256) 8e9b9fc162ad0476127630e4a6e806adb7a2883d24ce82da0bd8bb04b5d48ba2

⚡ AGENT INFO USK v3

Capabilities
knowledge_graph entity_extraction relation_detection graph_data memory_structuring
Permissions
✗ network
✗ filesystem
✗ subprocess
Interface
type: cli   entry_point: main.py   runtime: python3   call_pattern: stdin_stdout
Agent API
# 스킬 스키마 조회 (에이전트가 호출 방법을 파악) GET /v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/schema # 플랫폼별 자동 변환 다운로드 GET /v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=OpenClaw GET /v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=ClaudeCode GET /v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=ClaudeCodeAgentSkill GET /v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=Cursor GET /v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=GeminiCLI GET /v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=CodexCLI GET /v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=CustomAgent

Installation

Compatible Platforms any

1
Install the skill using openclaw_skill_manager.py.
python openclaw_skill_manager.py --install knowledge-graph-builder
2
Verify installation
python openclaw_skill_manager.py --list-installed
3
Install a specific version (optional)
python openclaw_skill_manager.py --install knowledge-graph-builder --version 1.0.0
1
Download the skill package.
curl -O https://aiskillstore.io/v1/skills/41c1bae7-c70d-4367-954a-5c5260283746/download
2
Place it in the Claude Code commands directory.
unzip knowledge-graph-builder.skill -d ~/.claude/commands/knowledge-graph-builder/
3
Use it as a slash command in Claude Code.
/knowledge-graph-builder
1
Download the Agent Skills package.
curl -O https://aiskillstore.io/v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=ClaudeCodeAgentSkill
2
Unzip it into the Claude Code skills directory.
unzip knowledge-graph-builder-agent-skill-*.skill -d ~/.claude/skills/knowledge-graph-builder/
3
Restart Claude Code — the skill is auto-loaded at session start. No slash command needed.
1
Download the Cursor-converted package.
curl -O https://aiskillstore.io/v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=Cursor
2
Unzip and place it in a permanent location.
unzip knowledge-graph-builder-cursor-*.skill -d ~/.cursor/skills/knowledge-graph-builder/
3
Add the MCP server config to .cursor/mcp.json, then restart Cursor.
cat ~/.cursor/skills/knowledge-graph-builder/cursor_mcp_config.json
1
Download the Gemini CLI-converted package.
curl -O https://aiskillstore.io/v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=GeminiCLI
2
Unzip and place it in a permanent location.
unzip knowledge-graph-builder-geminicli-*.skill -d ~/.gemini/skills/knowledge-graph-builder/
3
Add the MCP server config to ~/.gemini/settings.json, then restart Gemini CLI.
cat ~/.gemini/skills/knowledge-graph-builder/gemini_settings_snippet.json
1
Download the Codex CLI-converted package.
curl -O https://aiskillstore.io/v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/download?platform=CodexCLI
2
Unzip and place it in a permanent location.
unzip knowledge-graph-builder-codexcli-*.skill -d ~/.codex/skills/knowledge-graph-builder/
3
Add the MCP server config to ~/.codex/config.toml, then restart Codex CLI.
cat ~/.codex/skills/knowledge-graph-builder/codex_config_snippet.toml
1
Download the skill package via REST API.
GET https://aiskillstore.io/v1/skills/41c1bae7-c70d-4367-954a-5c5260283746/download
2
Place it in your agent platform's skills directory.
cp knowledge-graph-builder.skill ./skills/
3
Fetch platform-specific details via the Install Guide API.
GET https://aiskillstore.io/v1/skills/41c1bae7-c70d-4367-954a-5c5260283746/install-guide?platform=CustomAgent

Security Vetting Report

Vetting Result APPROVED

Findings: ["메타데이터 경고: 권장 필드 없음: 'requirements' (SKILL.md v2 권장)"]

✅ No security risks found.

AI Review Stage

Reviewer gemini Risk Level 🟢 Low Review Summary 외부 의존성 없이 텍스트에서 지식 그래프를 추출하는 스킬로, 선언된 권한과 코드 내용이 일치하며 보안 위험이 낮습니다.
Reasoning

1. **선언된 permissions과 실제 코드 일치 여부:** 스킬 메타데이터에서 `network: false`, `filesystem: false`, `subprocess: false`로 명확하게 선언되어 있습니다. 제공된 `main.py` 코드 스니펫에서 사용된 `json`, `re`, `sys`, `unicodedata` 모듈은 모두 Python 표준 라이브러리에 포함되며, 선언된 권한을 위반하는 외부 통신, 파일 시스템 접근 또는 서브프로세스 실행 기능을 제공하지 않습니다. `sys.stdin` 및 `sys.stdout` 사용은 CLI 인터페이스의 표준 동작으로, 파일 시스템 접근 권한 위반으로 간주되지 않습니다. 이는 'Zero external dependencies'라는 설명과도 일치합니다. 2. **악의적 목적의 코드 여부:** 제공된 코드 스니펫에는 악의적인 목적(데이터 탈취, 시스템 파괴 등)을 가진 코드가 발견되지 않았습니다. 정적 분석 결과에서도 `red_flags_found` 및 `obfuscation_warnings`가 없다고 보고되어, 코드에 명백한 악성 행위나 난독화가 없음을 시사합니다. 3. **선언되지 않은 외부 통신 여부:** `permissions.network`가 `false`로 설정되어 있으며, 코드에서 외부 네트워크 통신을 수행하는 어떠한 징후도 발견되지 않았습니다. 사용된 모듈 또한 네트워크 기능을 포함하지 않습니다. 4. **사용자 데이터 무단 수집/전송 여부:** 네트워크 및 파일 시스템 접근 권한이 없으므로, 사용자 데이터를 무단으로 수집하여 외부로 전송하거나 영구적으로 저장할 수 없습니다. 스킬의 목적은 입력 텍스트를 처리하여 지식 그래프 데이터를 출력하는 것이며, 이는 스킬의 의도된 기능 범위 내에 있습니다. 5. **코드 품질 및 목적 일치 여부:** 스킬의 설명, 입력/출력 스키마, 예시가 명확하게 정의되어 있으며, 제공된 코드 스니펫은 NLP 패턴 정의에 대한 깔끔하고 구조화된 접근 방식을 보여줍니다. 이는 스킬의 목적과 일치하며, 전반적인 코드 품질이 양호할 것으로 예상됩니다.

Version History

Version USK v3 Vetting Status Packaged At Downloads Changelog
v1.0.0 approved 2026-06-06 ⬇ 3 1.0.0: Initial release — stdlib-only, English + Korean bilingual entity/relation extraction
v1.0.0 pending 2026-06-06 ⬇ 0 1.0.0: Initial release — stdlib-only, English + Korean bilingual entity/relation extraction
v1.0.0 pending 2026-06-06 ⬇ 0 1.0.0: Initial release — stdlib-only, English + Korean bilingual entity/relation extraction

Examples 6

Representative input/output examples for this skill. Agents can use these to understand how to invoke the skill and what output to expect.

english_basic
# english# person# org# founded

Extract a person-founded-org relation from an English sentence

📥 Input
{
  "language": "en",
  "text": "Elon Musk founded SpaceX in 2002 in California."
}
📤 Output
{
  "edges": [
    {
      "confidence": 0.9,
      "evidence": "Elon Musk founded SpaceX in 2002 in California.",
      "relation": "founded",
      "source_id": "n0",
      "target_id": "n1"
    }
  ],
  "nodes": [
    {
      "aliases": [],
      "id": "n0",
      "label": "Elon Musk",
      "mentions": 1,
      "type": "PERSON"
    },
    {
      "aliases": [],
      "id": "n1",
      "label": "SpaceX",
      "mentions": 1,
      "type": "ORG"
    },
    {
      "aliases": [],
      "id": "n2",
      "label": "2002",
      "mentions": 1,
      "type": "DATE"
    },
    {
      "aliases": [],
      "id": "n3",
      "label": "California",
      "mentions": 1,
      "type": "LOC"
    }
  ],
  "stats": {
    "edge_count": 1,
    "input_char_count": 47,
    "language_detected": "en",
    "node_count": 4
  }
}
korean_basic
# korean# person# org

Extract a person-role-org relation from a Korean sentence

📥 Input
{
  "language": "ko",
  "text": "\uae40\ucca0\uc218\ub294 \ub124\uc774\ubc84\uc758 CTO\uc774\ub2e4."
}
📤 Output
{
  "edges": [
    {
      "confidence": 0.8,
      "evidence": "\uae40\ucca0\uc218\ub294 \ub124\uc774\ubc84\uc758 CTO\uc774\ub2e4.",
      "relation": "role_at",
      "source_id": "n0",
      "target_id": "n1"
    }
  ],
  "nodes": [
    {
      "aliases": [],
      "id": "n0",
      "label": "\uae40\ucca0\uc218",
      "mentions": 1,
      "type": "PERSON"
    },
    {
      "aliases": [],
      "id": "n1",
      "label": "\ub124\uc774\ubc84",
      "mentions": 1,
      "type": "ORG"
    }
  ],
  "stats": {
    "edge_count": 1,
    "input_char_count": 14,
    "language_detected": "ko",
    "node_count": 2
  }
}
alias_merge
# alias-merge# person# org

Multiple references to the same person collapsed into one node with aliases list

📥 Input
{
  "language": "en",
  "merge_aliases": true,
  "text": "Sam Altman leads OpenAI. Altman previously worked at Y Combinator. Sam is known for his vision."
}
📤 Output
{
  "edges": [
    {
      "confidence": 0.9,
      "evidence": "Sam Altman leads OpenAI.",
      "relation": "leads",
      "source_id": "n0",
      "target_id": "n1"
    },
    {
      "confidence": 0.75,
      "evidence": "Altman previously worked at Y Combinator.",
      "relation": "works_at",
      "source_id": "n0",
      "target_id": "n2"
    }
  ],
  "nodes": [
    {
      "aliases": [
        "Altman",
        "Sam"
      ],
      "id": "n0",
      "label": "Sam Altman",
      "mentions": 3,
      "type": "PERSON"
    },
    {
      "aliases": [],
      "id": "n1",
      "label": "OpenAI",
      "mentions": 1,
      "type": "ORG"
    },
    {
      "aliases": [],
      "id": "n2",
      "label": "Y Combinator",
      "mentions": 1,
      "type": "ORG"
    }
  ],
  "stats": {
    "edge_count": 2,
    "input_char_count": 99,
    "language_detected": "en",
    "node_count": 3
  }
}
entity_type_filter
# filter# person-only

Filter to extract only PERSON entities, ignoring ORG and LOC

📥 Input
{
  "entity_types": [
    "PERSON"
  ],
  "language": "en",
  "text": "Alice and Bob work at Google in New York."
}
📤 Output
{
  "edges": [],
  "nodes": [
    {
      "aliases": [],
      "id": "n0",
      "label": "Alice",
      "mentions": 1,
      "type": "PERSON"
    },
    {
      "aliases": [],
      "id": "n1",
      "label": "Bob",
      "mentions": 1,
      "type": "PERSON"
    }
  ],
  "stats": {
    "edge_count": 0,
    "input_char_count": 41,
    "language_detected": "en",
    "node_count": 2
  }
}
meeting_notes
# meeting-notes# actions# dates

Extract action owners and deadlines from meeting notes

📥 Input
{
  "language": "en",
  "text": "Alice will deliver the report by Friday. The review will be done by Bob before Monday."
}
📤 Output
{
  "edges": [
    {
      "confidence": 0.7,
      "evidence": "Alice will deliver the report by Friday.",
      "relation": "deadline",
      "source_id": "n0",
      "target_id": "n2"
    },
    {
      "confidence": 0.7,
      "evidence": "The review will be done by Bob before Monday.",
      "relation": "deadline",
      "source_id": "n1",
      "target_id": "n3"
    }
  ],
  "nodes": [
    {
      "aliases": [],
      "id": "n0",
      "label": "Alice",
      "mentions": 1,
      "type": "PERSON"
    },
    {
      "aliases": [],
      "id": "n1",
      "label": "Bob",
      "mentions": 1,
      "type": "PERSON"
    },
    {
      "aliases": [],
      "id": "n2",
      "label": "Friday",
      "mentions": 1,
      "type": "DATE"
    },
    {
      "aliases": [],
      "id": "n3",
      "label": "Monday",
      "mentions": 1,
      "type": "DATE"
    }
  ],
  "stats": {
    "edge_count": 2,
    "input_char_count": 87,
    "language_detected": "en",
    "node_count": 4
  }
}
empty_text_error
# error# validation

Empty or whitespace-only text returns a structured error

📥 Input
{
  "text": ""
}
📤 Output
{
  "error": {
    "code": "EMPTY_TEXT",
    "message": "Input text must not be empty. Provide at least one sentence."
  }
}

All examples are also available via the agent API: /v1/agent/skills/41c1bae7-c70d-4367-954a-5c5260283746/schema

Reviews & Ratings

No reviews yet. Be the first to leave one!

✍️ Write a Review