| name | reader |
| description | Local-first reading and distillation engine for documents and pasted text. Use whenever the user wants to read, summarize, extract key points, compare documents, understand a report, pull action items, or turn long text into a clear brief. Works locally with plain text, markdown, HTML, JSON, and other text-like files. No external sync or credentials. |
Helper Behavior Review
python scripts/_run.py is an optional helper whose behavior should be compared with this skill's documented purpose before use. Show what the script does, explain any mismatch with the user-facing description, and run it only when the user explicitly wants that diagnostic path.
Reader: Read less. Understand more.
Core Philosophy
- Distill before overwhelming.
- Prefer clarity over verbosity.
- Extract decisions, actions, and questions — not just summaries.
- Keep reading local, transparent, and low-friction.
Runtime Requirements
- Python 3 must be available as
python3
- No external packages required
Storage
All data is stored locally only under:
~/.openclaw/workspace/memory/reader/history.json
No external sync. No cloud storage. No third-party APIs.
Use Cases
- Summarize a document or pasted text
- Extract key points, action items, and open questions
- Compare two drafts or reports
- Turn long text into a concise reading brief
- Help the user understand dense writing faster
Key Workflows
- Read:
read_text.py --file path/to/file
- Summarize:
summarize.py --file path/to/file --style executive
- Brief:
extract_brief.py --file path/to/file
- Compare:
compare_texts.py --file_a old.txt --file_b new.txt
Scripts
| Script | Purpose |
|---|
init_storage.py | Initialize local storage |
read_text.py | Normalize and read local text-like files |
summarize.py | Generate concise summaries |
extract_brief.py | Extract summary, key points, actions, and questions |
compare_texts.py | Compare two documents |
