dont-be-greedy
CommunityPrevent context overflow with size checks.
AuthorElliotJLT
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This skill mitigates context overflow by measuring files before loading, chunking oversized data, and returning safe previews and summaries to keep downstream processes running smoothly.
Core Features & Use Cases
- Size-aware loading: measure file size and tokens before loading.
- Automatic chunking and per-chunk summarization for large files.
- Quick inspection and safe previews to aid quick decisions.
- Use Case: A data analyst receives a CSV or JSON dataset larger than the token limit and wants a compact summary plus representative snippets without overwhelming the model.
Quick Start
- Estimate token cost for a file: python scripts/estimate_size.py "<file_path>"
- Choose the action based on estimated tokens:
- If < 10,000: perform quick inspection and load directly
- If 10,000–30,000: perform quick inspection and consider filtering
- If > 30,000: chunk the file and summarize each chunk
- Run the appropriate workflow:
- For small files: python scripts/quick_inspect.py "<file_path>"
- For large files: python scripts/chunker.py "<file_path>" && python scripts/summarize.py "<chunk_file>"
- Return a structured output including overall summary, safe preview, and next steps.
Dependency Matrix
Required Modules
None requiredComponents
scripts
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: dont-be-greedy Download link: https://github.com/ElliotJLT/Claude-Skill-Potions/archive/main.zip#dont-be-greedy Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.