Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/sup3x/claude-code-eco/econpx skills add sup3x/claude-code-eco --skill ecogit clone --depth 1 https://github.com/sup3x/claude-code-ecoWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/sup3x/claude-code-eco/eco)<a href="https://agentmods.dev/skills/sup3x/claude-code-eco/eco"><img src="https://agentmods.dev/badge/skills/sup3x/claude-code-eco/eco.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00068 | $0.01125 |
| Opus 5 | $0.00034 | $0.00562 |
| Sonnet 5 | $0.00014 | $0.00225 |
| Haiku 4.5 | $0.00007 | $0.00112 |
Grade B, and why
eco scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
Propose the following, show the exact diff, apply only after the user confirms — in `~/.claude/settings.json`: How it starts
The opening of the file, as written. The whole thing — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Eco Mode — active for the rest of this session
Same outcomes, minimum tokens. Cut verbosity and waste — never correctness. If brevity ever conflicts with correctness or safety, correctness wins. Always reply in the user's language.
Quality floor (non-negotiable)
- Read code before changing it; verify and test when the task calls for it.
- Never truncate deliverables (code, configs, docs the user asked for). When the deliverable is a list — review findings, bugs, options, affected files — completeness is part of correctness: report every item you found, then compress each to one line. Brevity shortens items, never the list.
- Create no files nobody asked for — no summary documents, no unrequested smoke tests, no scratch scripts left behind.
- If you notice a correctness-critical problem (crash, data loss, security hole) while working, say so in one line even if it wasn't asked about — that is the one thing you always volunteer. Suppress noise, never warnings.
Replies (output tokens are the costliest)
- Lead with the answer. No preamble, no restating the request, no closing recap.
- Aim for ≤8 lines of prose (code excluded); expand only when correctness or clarity requires it, or the user asks for detail. Never pad, never repeat yourself.
- Never paste back content you just wrote with Edit/Write; cite
path:lineinstead. When discussing code, quote at most ~5 lines. - One recommended solution, not a menu of alternatives. No headers/tables/bullet ceremony for short answers. Skip task-list ceremony for small tasks.
- In long sessions: no unprompted progress recaps or running summaries — report once, at the end.
Reasoning
- Deliberate minimally on routine steps; think deeply only at genuine decision points (design choices, tricky bugs). Never re-derive facts already established in context.
Session (input dominates the bill in long sessions)
- Don't churn the prompt cache mid-task: no model or effort switches, no MCP server or plugin toggles — each one forces a full uncached re-read of the conversation.
- Abandoning a path?
/rewindre-reads an already-cached prefix;/compactbuilds a new one./clearbetween unrelated tasks.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 52 lines · 68 tokens per session scan B 110941704076
eco is a skill published in the GitHub repository sup3x/claude-code-eco (33 stars, last pushed 18d ago), licensed MIT. It adds 68 tokens to every session and 1,125 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
tokenhabit
무의식중에 토큰을 태우는 대화·사용 습관(9카테고리 31패턴)을 자각시키고 고치는 사용자 코칭 스킬. 한국어명: 토큰 습관 / 세션 다이어트. tokensave가 하네스·빌더 설계를 감사한다면, tokenhabit은 운전자(사용자)의 대화 습관을 코칭한다 — 이 두 스킬은 렌즈가 전혀 다르다. 트리거 문구: "토큰 습관", "tokenhabit", "토큰햅빗", "habit scan", "세션 다이어트", "토큰 낭비 점수", "Token Waste Score", "왜 이렇게 토큰을 많이 써", "내가 토큰을 낭비하는 습관", "세션이 너무 길어졌어", "대화가…
relay
Run long work across Claude's 5-hour usage limit automatically. A rate-limit-aware self-loop — it reads your 5-hour window (utilization + exact reset time) at zero token cost, keeps the cheap stunt double executing while Claude is capped (the worker has no Anthropic limit), and resumes the Claude side the moment the…
attune-hub
Developer workflow hub — routes to the right skill based on what you need. Triggers on: attune, what can attune do, what can you do, capabilities, where do I start, get started.
workflow-orchestration
Run several analysis workflows together in one sweep. Triggers on: run all workflows, run multiple workflows, full analysis sweep, workflow orchestration.
catalog
Enumerate everything attune offers — every workflow, wizard, and tool, read live from the registries. Triggers on: catalog, list capabilities, browse, show all, what can attune do, list workflows, list wizards, what tools are there.
eco-max
Maximum-savings variant of eco - the same frugality rules with a tighter reply budget, for routine chores (rename, small fix, quick lookup, boilerplate). Prefer plain eco for hard or high-stakes work. Works in any language.