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...

Security Scan

This skill's code and instructions are coherent with its stated purpose: it reads an Ontology graph.jsonl and optional schema to generate a filesystem ExpertPack; it does not request unrelated credentials, network access, or privileged system changes.

Purpose & Capability

Name/description match the behavior: the SKILL.md and convert.py both target conversion of an Ontology graph.jsonl (and optional schema.yaml) into an ExpertPack. Required binary is only python3, which is appropriate for the included Python script.

Instruction Scope

Runtime instructions explicitly tell the agent to read the Ontology skill's graph.jsonl and optional schema.yaml and write files to a user-specified output directory. This is within scope. Note: the converter writes files under the chosen output path and attempts to sanitize top-level properties that look like credentials, but it only inspects property keys (not arbitrary nested content), so sensitive data embedded in values or nested structures could still be copied into the generated pack.

Install Mechanism

Instruction-only skill with an included Python script and no install spec. No external downloads, package installs, or scripts fetched from remote URLs — lowest-risk install model.

Credentials

The skill requests no environment variables, no credentials, and no config paths. The only filesystem access is to the graph and schema paths the user provides and to the output directory — appropriate for its purpose.

Persistence & Privilege

always is false and the skill does not request persistent or system-wide privileges. It writes output only to the user-specified output directory and does not modify other skills or global agent settings.

Assessment

This skill appears to do exactly what it says: convert a graph.jsonl into an ExpertPack using the included Python script. Before running: (1) Verify the graph.jsonl and optional schema.yaml paths you pass in are the intended files (don't point to unrelated system or secret stores). (2) Review generated files before sharing externally — the script removes properties whose keys look like credentials but may not catch secrets embedded in values or nested structures. (3) Run the tool in a directory you control (or a disposable environment) and commit/back up your original graph if you need to preserve provenance. If you need stronger secret-scrubbing, inspect and extend sanitize_properties to handle nested structures or pattern-match secret-like values.

Ontology to ExpertPack Converter

Converts an OpenClaw Ontology skill's append-only knowledge graph into a fully compliant ExpertPack with multi-layer retrieval support.

How to Use

Run the converter script:

python3 {skill_dir}/scripts/convert.py \
  --graph memory/ontology/graph.jsonl \
  --output ~/expertpacks/my-knowledge-pack

Optional flags:

• --schema memory/ontology/schema.yaml — uses type definitions and relation rules
• --name "My Knowledge Pack" — custom pack name (defaults to "Ontology Export")
• --type auto|person|product|process|composite — override auto-detected pack type

What It Produces

A complete ExpertPack at the output directory:

• manifest.yaml — pack identity, type, context tiers, EK metadata placeholder
• overview.md — summary of graph contents, entity/relation counts, navigation guide
• Content organized by mapped category (relationships/, workflows/, facts/, concepts/, operational/, governance/)
• _index.md in each content directory
• relations.yaml — typed entity relation graph (schema 2.3 compliant)
• glossary.md — entity types and terms
• Lead summaries and ## section headers for optimal chunking

Filenames use kebab-case. Content files kept under 3KB.

Post-Conversion Steps

1. cd into the generated ExpertPack directory
2. Run the ExpertPack chunker to generate summaries/ and propositions/
3. Run EK evaluator to measure esoteric knowledge ratio
4. Review and refine manifest.yaml context tiers
5. Commit to git and share via expertpack.ai or ClawHub

See expertpack.ai and the expertpack ClawHub skill for full pack maintenance workflows.

Keep the output pack git-friendly and ready for iterative deepening.