Dont Be Greedy

When a user uploads or references a data file (CSV, JSON, XLSX, TXT, LOG) or any file larger than 100KB, immediately estimate token cost using scripts/estimate_size.py. If >30k tokens, chunk the file and summarize each chunk. If smaller, run quick inspection. Return a safe preview and summary without asking the user what to do.

62 stars 2 forksDocuments & Office Files

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add ElliotJLT/Claude-Skill-Potions --skill dont-be-greedy
Manual install
git clone --depth 1 https://github.com/ElliotJLT/Claude-Skill-Potions.git /tmp/Claude-Skill-Potions
mkdir -p ~/.claude/skills
cp -r /tmp/Claude-Skill-Potions/skills/dont-be-greedy ~/.claude/skills/dont-be-greedy

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Don't Be Greedy

<purpose>
Prevents context overflow by enforcing size-aware data loading. Large files can
exceed context windows and crash agent workflows. This skill measures files before
loading, chunks oversized data, and returns compact summaries with safe previews
so downstream processing can continue without context exhaustion.
</purpose>

## Instructions

### Step 1: Estimate Token Cost

Before loading ANY data file:

```bash
python scripts/estimate_size.py "<file_path>"
```

This returns byte count and estimated token count.

### Step 2: Apply Strategy Based on Size

| Estimated Tokens | Action |
|------------------|--------|
| < 10,000 | Run quick inspection, load directly |
| 10,000 - 30,000 | Run quick inspection, consider filtering |
| > 30,000 | Chunk and summarize before loading |

### Step 3: Execute Appropriate Workflow

<strategy name="small-file">
For files under 10k tokens:

```bash
python scripts/quick_inspect.py "<file_path>"
```

Return stats and load file directly.
</strategy>

<strategy name="large-file">
For files over 30k tokens:

```bash
python scripts/chunker.py "<file_path>"
python scripts/summarize.py "<chunk_file>"
```

Return overall summary + per-chunk summaries + safe preview of first rows.
</strategy>

### Step 4: Return Structured Output

Always provide:
- Overall summary (1-3 paragraphs)
- Safe preview (first N rows/lines)
- Recommendation for next steps
- Chunk information if file was split

## NEVER
Read the full skill on GitHub

Details

Skill name
dont-be-greedy
Path
skills/dont-be-greedy/SKILL.md
License
MIT
Repo updated
12 February 2026

FAQ

What does the Dont Be Greedy skill do?

When a user uploads or references a data file (CSV, JSON, XLSX, TXT, LOG) or any file larger than 100KB, immediately estimate token cost using scripts/estimate_size.py. If >30k tokens, chunk the file and summarize each chunk. If smaller, run quick inspection. Return a safe preview and summary without asking the user what to do.

How do I install the Dont Be Greedy skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add ElliotJLT/Claude-Skill-Potions --skill dont-be-greedy. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Dont Be Greedy popular?

It lives in ElliotJLT/Claude-Skill-Potions, which has 62 GitHub stars and 2 forks.

Related skills

Agent Audit

62

ElliotJLT/Claude-Skill-Potions

When reviewing, optimizing, or inheriting an LLM agent, systematically identify what's actually load-bearing vs. cruft. Instruments usage, ablates components, and strips the agent to minimum viable form. Answers: "What can we remove without degrading quality?" The goal is the leanest agent that still works.

AI, LLMs & Agents

Battle Plan

62

ElliotJLT/Claude-Skill-Potions

Before ANY significant development task (new feature, refactor, integration, migration), run a complete planning ritual by orchestrating other skills in sequence: rubber-duck (clarify scope) -> pre-mortem (assess risks) -> eta (estimate time) -> final confirmation. Do not start coding until the battle plan is approved.

Software Development

Breadcrumbs

62

ElliotJLT/Claude-Skill-Potions

Leaves notes for future Claude sessions in .claude/breadcrumbs.md. At session end or after significant work, records what was tried, what worked, what failed, and context for next time. On session start, reads breadcrumbs to understand history. Session-to-session memory without manual updates.

Productivity & Notes

Lit Review Table

36

dariia-m/my_claude_skills

Build an Excel literature table on a topic. Searches economics journals and working papers (via OpenAlex) and compiles a formatted .xlsx with columns: authors, year, title, journal, full citation (AEA), keywords, abstract, citation count, paper type, OA status, link - plus any custom columns the user asks for in that run. Optionally downloads the papers' PDFs to a folder (open-access/working-paper version, else marks "not available"). Use when the user asks to make a literature table or spreadsheet of papers, do a lit search into Excel, build a bibliography table, or collect/download papers on

Documents & Office Files11

Cn Comps Analysis

43

ccq1/cn-financial-services-plugins

构建机构级 A 股可比公司分析,包含经营指标、估值倍数和统计基准,输出 Excel 格式。 **适用场景:** - A 股上市公司估值(并购、投资分析) - 对标行业同行的经营表现 - 识别估值异常(高估/低估) - 支持投资决策报告 - 行业横向比较 **不适用:** - 未上市或无可比同行的公司 - 高度多元化的集团公司 - ST/*ST 退市风险公司 - 尚无营收的初创公司

Documents & Office Files

Resuming Work

33

okgrit/done-is-a-claim

Resume a saved task after a session, context or machine change by checking its checkpoint, artifacts and current ownership. Use when receiving unfinished work or reconciling a stop; ordinary uninterrupted edits do not need a handoff.

Documents & Office FilesIn 1 list

Browse the most popular Claude skills, all documents & office files skills, or write your own with the skills generator.