OpenParable
Skip to content
← Back to skills

Extract Entities

Documents
extract-entities

Extract Entities is an Agent Skill for documents published in the superlinked/sie repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Extract people, organizations, dates, amounts, or custom labels from a document through the connected Superlinked MCP edge, returning a compact table instead of reading the full document into context. Use when the user asks to list, extract, or tabulate entities from a file.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/superlinked/sie/blob/HEAD/packages/sie_mcp/plugin/claude-code/extract-entities/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/extract-entities, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Extract Entities with Superlinked MCP

Use the connected Superlinked MCP server to run zero-shot entity extraction on the SIE cluster. This mirrors the extract-entities flow from the sie_tools plugin in PR #1336, but uses the MCP extract_entities tool.

Steps

  1. Pick labels from the user's request. Lowercase nouns work best, for example: person, organization, date, amount, contract term.

  2. Do not open, view, attach, or read the whole source document into context.

  3. Call the Superlinked MCP extract_entities tool:

    • Pass labels as a list of label strings.
    • For PDF, scan, image-of-text, DOCX, PPTX, XLSX, or HTML inputs, read raw bytes only, base64-encode them, and pass document_base64 plus filename.
    • For text or markdown inputs, read raw bytes only, base64-encode them, and pass content_base64.
    • Leave engine as auto unless the user explicitly asks to force a conversion engine.

  4. Write the returned markdown_table to processed/<source-stem>.entities.md. Create processed/ if it does not exist.

  5. Your next message must begin with a short receipt:

    Extracted <entity_count> entities from <source> -> processed/<source-stem>.entities.md

    Then present the entity table or the subset the user asked for.

Errors

If the MCP call fails, relay the tool error and do not retry blindly. For auth or connector errors, tell the user to reinstall the generated MCP plugin pack or rerun the generated claude mcp add ... --header 'Authorization: Bearer ...' command from INSTALL.md.