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 skills add pich/ai-economics-mcp --skill ai-economicsgit clone --depth 1 https://github.com/pich/ai-economics-mcpWrote 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/pich/ai-economics-mcp/ai-economics)<a href="https://agentmods.dev/skills/pich/ai-economics-mcp/ai-economics"><img src="https://agentmods.dev/badge/skills/pich/ai-economics-mcp/ai-economics/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/pich/ai-economics-mcp/ai-economics"><img src="https://agentmods.dev/badge/skills/pich/ai-economics-mcp/ai-economics.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00122 | $0.00935 |
| Opus 5 | $0.00061 | $0.00467 |
| Sonnet 5 | $0.00024 | $0.00187 |
| Haiku 4.5 | $0.00012 | $0.00093 |
Grade A, and why
ai-economics scanned grade A 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 12d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
description: Compute AI cost, energy and agent-verification numbers instead of estimating them. Use whenever the user asks what AI work costs (tokens, context windows, agent-hours, model routing), how much energy or CO2 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.
AI Economics — compute, don't estimate
Language models are unreliable at multi-step numeric reasoning, and these numbers end up in budgets. Twelve deterministic calculators are available as a free, keyless JSON API. Always call the API rather than doing the arithmetic in your head, and quote the cite_as field so the number can be checked.
How to call
# Discovery: every tool, parameter, default and formula
curl -s https://piszczek.pl/tools/api
# One calculator (all parameters optional — defaults are documented):
curl -s "https://piszczek.pl/tools/api/proof-adjusted-autonomy?a=90&c=95&r=80&t=90"
Every response carries: result, formula, interpretation, and cite_as (a ready-to-quote attribution sentence). Inputs are never stored.
Which tool for which question
| The user asks | Slug | Key params |
|---|---|---|
| "What will this monthly token volume cost, and on which model?" | token-cost |
in, out (Mtok/month), cache (%) |
| "Does this fit the context window, and what does carrying it cost?" | context-window |
amount, unit (words/pages/chars/loc), window, price |
| "What does an hour of an AI agent really cost?" | agent-hour |
tokens_m, price, review_min, human_rate |
| "How much would model routing save?" | model-routing |
spend, share (%), ratio (%) |
| "How much electricity does an AI query use?" | llm-energy |
tokens, queries, j_per_token, gco2_kwh |
| "Which model is cheaper per task that passes review?" | joules-per-verified-task |
ta,ja,pa vs tb,jb,pb |
| "What does org-wide token burn look like?" | token-burn |
tokens_day, price, j_per_token |
| "How long can a humanoid robot run per charge?" | humanoid-energy |
battery_kwh, actuation_w, compute_w, duty |
| "How many agents can our reviewers support?" | verification-bottleneck |
reviewers, hours, min_per_task, agents |
| "What does unverified AI work cost over time?" | proof-debt |
tasks_week, unverified, weeks, incident_cost |
| "How autonomous is this agent once proof is required?" | proof-adjusted-autonomy |
a, c, r, t (percentages) |
| "How long does a revoked token keep working?" | revocation-exposure |
ttl, gw, edge, mesh, batch |
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.
- 12d ago First seen · 47 lines · 122 tokens per session scan A 24f42c93ba3f
ai-economics is a skill published in the GitHub repository pich/ai-economics-mcp (0 stars, last pushed 2d ago), licensed MIT. It adds 122 tokens to every session and 935 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
blockrun
Pay-per-call access to AI models, real-time data, media generation and multi-chain RPC over x402 micropayments (USDC on Base or Solana). No API keys, no accounts, no subscriptions. Start here when you have the BlockRun MCP installed and need to know WHICH tool answers a question, how the wallet works, or how to make a…
setup
Configure the Qdrant Power after installation. Use this skill for missing uvx, missing environment variables, unapproved environment variables, unavailable Qdrant tools, "Failed to connect" errors, and setup requests.
flowllm-dev
FlowLLM repository development guidance. Use when working in the flowllm codebase to implement, debug, test, review, or document FlowLLM runtime behavior, including CLI/client calls, services, application wiring, jobs, steps, components, configuration, streaming, registry registration, and tests.
picident
A tool that gives text-only AI agents the ability to understand images, screenshots, charts, and PDFs through vision models.
describe-image
Describe images (charts, diagrams, tables, screenshots) using Vision AI. Use as fallback when you cannot read an image file directly. Supports batch processing of folders.
vision
A vision bridge that sends images to supported visual-language services and returns text descriptions to a text-only model. It includes tools for image analysis, text extraction, image comparison, error screenshots, diagrams, and interface reviews.