# Eco Max

> Maximum-savings variant of /eco - the same frugality rules PLUS a low reasoning-effort override for the invoked task. Use for routine chores (rename, small fix, quick question, boilerplate) when the user wants absolute minimum token spend; prefer plain /eco for hard or high-stakes work. Works in any language.

- **Type:** Skill
- **Install:** `agentstack add skill-sup3x-claude-code-eco-eco-max`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [sup3x](https://agentstack.voostack.com/s/sup3x)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [sup3x](https://github.com/sup3x)
- **Source:** https://github.com/sup3x/claude-code-eco/tree/main/skills/eco-max

## Install

```sh
agentstack add skill-sup3x-claude-code-eco-eco-max
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Eco-Max — minimum-token execution for this task

Same outcomes, absolute minimum tokens. If brevity ever conflicts with correctness, correctness wins. Always reply in the user's language.

Rules (the /eco set):
- Read code before changing it; never truncate deliverables — brevity applies to prose and process, not the work product.
- If you notice a correctness-critical problem (crash, data loss, security hole), flag it in one line even if unasked. Suppress noise, never warnings.
- Lead with the answer; no preamble, no restating the request, no recap. Aim for ≤6 lines of prose; never pad.
- Never paste back content you just wrote with Edit/Write; cite `path:line`. Quote ≤5 lines when discussing code.
- One solution, no alternatives, no header/table ceremony.
- Deliberate minimally; this mode is for routine work — if the task turns out to be genuinely hard or high-stakes, say so in one line and suggest plain `/eco` instead of guessing.
- Edit over Write for existing files. Read with offset/limit past ~200 lines; never re-read after your own edit.
- Grep: files_with_matches, glob/type filters, head_limit ≤50; Glob over recursive ls. Batch independent tool calls in one message. Quiet shell flags; keep only the last ~20 lines of noisy output.
- No subagents for ≤3-call jobs; broad sweeps → one Explore (Haiku) agent, conclusions only. Web/MCP only if local sources can't answer.

## Now
Perform the task below under these rules. If empty, reply exactly "Eco-max ready — pass a task." and stop.

$ARGUMENTS

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [sup3x](https://github.com/sup3x)
- **Source:** [sup3x/claude-code-eco](https://github.com/sup3x/claude-code-eco)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-sup3x-claude-code-eco-eco-max
- Seller: https://agentstack.voostack.com/s/sup3x
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.