← 返回 Skills 市场

Elite To Expertpack

Name: Elite To Expertpack
Author: brianhearn

作者 Brian Hearn · GitHub ↗ · v1.0.1 · MIT-0

cross-platform ✓ 安全检测通过

268

总下载

当前安装

版本数

在 OpenClaw 中安装

/install elite-to-expertpack

功能描述

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...

使用说明 (SKILL.md)

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

cd ~/expertpacks/my-agent-pack
Verify content files are 400–800 tokens each (Schema 2.5 — retrieval-ready by design)
Measure EK ratio: python3 /path/to/expertpack/tools/eval-ek.py .
Review overview.md and manifest.yaml
Commit to git and publish to ClawHub

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

See also: Elite Longterm Memory skill on ClawHub.

安全使用建议

This skill appears coherent and runs only on your local files, but review these before using: 1) Run the script in a test environment or on a copy of your workspace first (it writes output). 2) Ensure python3 and PyYAML are installed (the script will exit with an instruction to pip install pyyaml). 3) Inspect the produced ExpertPack and overview.md to confirm secret redaction worked (the script uses regexes but regexes can miss edge cases). 4) The source is listed as unknown — if you care about provenance, verify the code yourself or obtain it from a trusted source (homepage is provided). 5) No network calls are present in the script, but if you later publish the pack (steps mention ClawHub) be mindful of any sensitive data you might be uploading. If you want higher assurance, have someone you trust review scripts/convert.py before running it on sensitive agent data.

功能分析

Type: OpenClaw Skill Name: elite-to-expertpack Version: 1.0.1 The elite-to-expertpack skill is a utility designed to migrate agent memory data between different storage formats. The Python script (scripts/convert.py) performs local file operations to read session states, journals, and curated archives, and includes a proactive security feature to identify and redact sensitive API keys (OpenAI, GitHub, Slack) using regular expressions. The instructions in SKILL.md are consistent with the code's functionality, and there is no evidence of data exfiltration, unauthorized execution, or malicious intent.

能力评估

✓ Purpose & Capability

The name/description promise (migrating the Elite 5-layer memory into an ExpertPack) matches what the Python script and SKILL.md do: they parse SESSION-STATE.md, MEMORY.md, daily journals, topic files, and Git-Notes JSONL and produce a structured output. There are no unrelated credential or network requirements declared.

✓ Instruction Scope

The SKILL.md instructs running the included script against a specified workspace and output path; the script only reads local files under the user's workspace (e.g., ~/.openclaw/*, memory/*.md, Git-Notes JSONL) and writes the ExpertPack to the output directory. It also performs secret-stripping by regex. The instructions do not request or read unrelated system configuration, environment variables, or external endpoints.

ℹ Install Mechanism

There is no install spec (instruction-only), which is lowest-risk. The script requires python3 (declared) and PyYAML (import yaml). PyYAML is not listed in declared requirements; the script exits with an error message instructing the user to pip install pyyaml. This is a minor operational gap but not a security red flag.

✓ Credentials

The skill requires no environment variables, credentials, or config paths beyond reading the user's local Elite memory files. The secret-stripping regexes are appropriately applied to inputs. No broad or unrelated credentials are requested.

✓ Persistence & Privilege

The skill is not marked always:true and does not attempt to modify other skills or system-wide settings. It runs on demand and writes output only to the specified output directory.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install elite-to-expertpack
安装完成后，直接呼叫该 Skill 的名称或使用 /elite-to-expertpack 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.1

Core 2.8: Obsidian compatibility — output packs include YAML frontmatter and can be opened as Obsidian vaults.

v2.0.0

Updated for Schema 2.7. Volatile/ directory support for time-bound EK migration.

v1.2.0

Removed chunker references. Schema 2.5 file-size verification step.

v1.1.0

Schema 2.5: Chunker step marked as legacy/optional. New packs authored to spec need no external chunker.

v1.0.0

Initial release — converts Elite Longterm Memory 5-layer system (SESSION-STATE, MEMORY.md, daily journals, Git-Notes JSONL, topic files) into structured ExpertPacks with context tiers, relations.yaml, and deduplication.

元数据

Slug elite-to-expertpack

版本 1.0.1

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 5

常见问题