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/eco-reportnpx skills add sup3x/claude-code-eco --skill eco-reportgit 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-report)<a href="https://agentmods.dev/skills/sup3x/claude-code-eco/eco-report"><img src="https://agentmods.dev/badge/skills/sup3x/claude-code-eco/eco-report.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 | $0.00062 | $0.00632 |
| Opus 5 | $0.00031 | $0.00316 |
| Sonnet 5 | $0.00012 | $0.00126 |
| Haiku 4.5 | $0.00006 | $0.00063 |
Grade A, and why
eco-report scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Eco Report - measured token accounting
Every number comes from the local JSONL transcript store. Read them; never estimate them.
Run it
node "${CLAUDE_PLUGIN_ROOT}/scripts/report.mjs"
If ${CLAUDE_PLUGIN_ROOT} did not expand (personal or project install rather than a plugin), run
report.mjs from this skill's own directory - the installer stages it there - and fall back to
scripts/report.mjs in the claude-code-eco checkout.
Pass the user's arguments straight through:
--days <n>window, default 7,0for all history--limit <n>session rows to print (totals always cover every session)--dir <path>a transcript store other than$CLAUDE_CONFIG_DIR/projects--jsonmachine-readable output
Present it
- The script's table is already on screen as the command's output. Do not reprint it - it is a finished report, and repeating it burns tokens on something the user is already looking at. Do not re-format it, re-order it, re-tally it, or drop lines to save space.
- Then add at most three lines of your own: the biggest cost driver visible in the table and one concrete action (for example a session with a high out/turn, or a cache write/read ratio above ~0.3, which means the prompt prefix keeps being rebuilt).
- Keep the "Observational, NOT a controlled A/B" disclaimer. The eco-armed vs not gap is not a
measured saving - the user chose when to run
/eco, so difficulty is a confound. State it as a correlation or not at all; the controlled numbers live inbenchmarks/results.md. - Never quote a cost in currency. Transcripts record tokens, not prices. If asked for money, say the price is not in the data and point at the published rates for the models listed.
turnsare API requests, deduped byrequestId;thinkingis a subset ofoutput, not an addition to it. Say so if the user reads the columns as additive.- On "No sessions ...", the window is simply empty: offer
--days 0. Do not go hunting through other files.
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.
- 5d ago First seen · 47 lines · 62 tokens per session scan A 2a80632f644a
eco-report is a skill published in the GitHub repository sup3x/claude-code-eco (33 stars, last pushed 19d ago), licensed MIT. It adds 62 tokens to every session and 632 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
lynkr
Universal LLM gateway with intelligent routing, Graphify code intelligence, Distill compression, routing telemetry, Code Mode, and 12+ provider support. 60-80% cost reduction for Claude Code, Cursor, and Codex.
uncommonroute-release
Use when publishing UncommonRoute. A release is only complete after all required steps are done: version sync, validation, GitHub push/tag/release, PyPI publish, and npm publish.
model-router
Use when route AI model requests to the optimal provider based on task, cost, latency, and capability requirements. Manage multi-provider LLM deployments. Use when working with model router.
elicit
Form-driven Socratic discovery — batch independent decision dimensions into one multi-select-capable form instead of N button-turns. Triggers on: scope this, discovery form, ask me everything at once, multi-select question.
roundtable
Convene the multi-LLM round table — Claude, Antigravity, and Codex deliberate a question; the user chairs promotion. Triggers on: roundtable, round table, convene the table, ask the table, what do the other models think, deliberate.
spec
Spec Ladders — goal-driven spec development: brainstorm, plan, review, and execute a gated task ladder with recorded approvals. Triggers on: spec, spec ladders, brainstorm and build, plan and execute, idea to code, build from scratch.