Elite to ExpertPack

Convert Elite Longterm Memory data into a structured ExpertPack. Migrates the 5-layer memory system (SESSION-STATE hot RAM, LanceDB warm store, Git-Notes col...

Security Scan

The skill's code, instructions, and requirements are coherent with its stated purpose (converting local Elite Longterm Memory files into an ExpertPack) and do not request unrelated credentials or perform network installs.

Purpose & Capability

Name and description match the included script and SKILL.md: the converter reads local Elite memory files (SESSION-STATE.md, MEMORY.md, memory/*.md, Git-Notes JSONL, optional LanceDB path) and emits an ExpertPack. Requested binaries (python3) and no environment variables are proportional to this purpose.

Instruction Scope

The SKILL.md instructs the user to run the included python script against a workspace and write output to a chosen directory. The script reads files under the provided --workspace and (optionally) default locations such as ~/.openclaw/memory/lancedb. It strips secrets using regexes and warns about skipped cloud stores. This is in-scope for migration, but be aware it will read local user files (including home-dir paths) and write output files — review outputs before publishing.

Install Mechanism

No install spec is provided and the skill is instruction-only with a single Python script. No remote downloads or package installations are performed by the skill itself. The script does require PyYAML to be present and exits with an error if it's missing.

Credentials

The skill requests no environment variables, no credentials, and no special config paths beyond reading files in the user-specified workspace and the documented default LanceDB path. The secret-stripping regexes exist in the script but no sensitive credentials are required by the skill.

Persistence & Privilege

always is false, there is no install that persists or modifies other skills, and the skill does not attempt to change system-wide agent settings. It only writes the output ExpertPack to the specified output directory.

Assessment

This converter is coherent and runs locally with Python. Before running: 1) Inspect the output directory after conversion and before publishing — the script attempts to redact secrets using regexes but regex-based redaction is not guaranteed to find every secret. 2) Provide an explicit --workspace and --output paths that limit scope (avoid running on an entire home directory unintentionally). 3) The script may read defaults such as ~/.openclaw/memory/lancedb if present; if you do not want that read, point --workspace to a safer path or remove/snapshot files. 4) Ensure PyYAML is installed in a trusted environment; the script will exit if missing. 5) The SKILL.md references additional ExpertPack tools (chunker, eval-ek) that are not included here — verify those tools separately before invoking them. Overall the skill appears to do what it claims, but exercise normal caution with any tool that reads and aggregates local documents and journals.

Elite Longterm Memory → ExpertPack

Converts an Elite Longterm Memory (5-layer system with 32K ClawHub downloads) into a proper structured ExpertPack.

Supported layers:

• Hot RAM — SESSION-STATE.md (current task, context, decisions)
• Warm Store — LanceDB vectors at ~/.openclaw/memory/lancedb/ (note: exported or skipped)
• Cold Store — Git-Notes JSONL (decisions, learnings, preferences)
• Curated Archive — MEMORY.md, memory/YYYY-MM-DD.md journals, memory/topics/*.md
• Cloud — SuperMemory/Mem0 (skipped, noted in overview)

Usage

cd /root/.openclaw/workspace/ExpertPack/skills/elite-to-expertpack
python3 scripts/convert.py \
  --workspace /path/to/your/workspace \
  --output ~/expertpacks/my-agent-pack \
  [--name "My Agent's Knowledge"] \
  [--type auto|person|agent]

Flags let you override auto-detected paths for each layer.

What It Produces

A complete ExpertPack conforming to schema 2.3:

• manifest.yaml (with context tiers, EK stub)
• overview.md summarizing conversion (layer counts, warnings)
• Structured directories: mind/, facts/, summaries/, operational/, relationships/, etc.
• _index.md files, lead summaries, glossary.md (if terms found)
• relations.yaml (if relationships detected)
• Clean deduplication preferring curated > structured > raw sources

Secrets are automatically stripped (sk-, ghp_, tokens, passwords). Warnings emitted for any found.

Post-Conversion Steps

1. cd ~/expertpacks/my-agent-pack
2. Run the ExpertPack chunker: python3 /path/to/expertpack/scripts/chunk.py --pack . --output ./.chunks
3. Measure EK ratio: python3 /path/to/expertpack/scripts/eval-ek.py .
4. Review overview.md and manifest.yaml
5. Commit to git and publish to ClawHub

Learn more: https://expertpack.ai • ClawHub expertpack skill

See also: Elite Longterm Memory skill on ClawHub.