Ontology To Expertpack

Convert an Ontology skill knowledge graph into a structured ExpertPack. Use when migrating from the Ontology skill's entity/relation graph (memory/ontology/g...

This skill appears to do exactly what it claims: read a local graph.jsonl (and optional schema.yaml) and produce an ExpertPack in the specified output directory. Before running: 1) point --graph at a copy of your ontology file (don't run on system-wide paths you don't intend to export); 2) review that sanitization behavior (the script omits any property whose key contains substrings like 'password', 'token', 'api_key', etc.) — if you need to preserve some fields with those substrings in their names, back them up first; 3) the script writes files to the chosen output directory (it will create folders and overwrite files with the same names), so run it where you want those files created; 4) the script has no networking or credential requirements, but inspect your graph.jsonl for sensitive data before exporting and committing the generated pack to remote repositories. If you want extra assurance, run the converter on a small sample of the graph first and inspect the output structure and content.

Type: OpenClaw Skill Name: ontology-to-expertpack Version: 1.0.1 The skill bundle is a utility for converting knowledge graphs from OpenClaw's Ontology format to the ExpertPack format. The Python script (scripts/convert.py) performs local file processing, includes a 'sanitize_properties' function to explicitly strip sensitive keys (e.g., passwords, tokens) during conversion, and uses safe filename sanitization (to_kebab) to prevent path traversal. No network activity, persistence mechanisms, or malicious instructions were found.