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 mcasillas17/mexican-mom --skill ni-que-fueramos-ricosgit clone --depth 1 https://github.com/mcasillas17/mexican-momWrote 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/mcasillas17/mexican-mom/ni-que-fueramos-ricos)<a href="https://agentmods.dev/skills/mcasillas17/mexican-mom/ni-que-fueramos-ricos"><img src="https://agentmods.dev/badge/skills/mcasillas17/mexican-mom/ni-que-fueramos-ricos/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/mcasillas17/mexican-mom/ni-que-fueramos-ricos"><img src="https://agentmods.dev/badge/skills/mcasillas17/mexican-mom/ni-que-fueramos-ricos.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 | $0.00077 | $0.01218 |
| Opus 5 | $0.00039 | $0.00609 |
| Sonnet 5 | $0.00015 | $0.00244 |
| Haiku 4.5 | $0.00008 | $0.00122 |
Grade A, and why
ni-que-fueramos-ricos 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 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.
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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
¡Apaga la luz, ni que fuéramos ricos!
The light is on in a room nobody is in. The money is not the point — the waste is.
Rule
Measure or bound the expensive path before you optimize it. You do not get to guess which line is hot; the line you would guess is usually not the one paying. Once you know what the path costs, remove the work that is genuinely waste — work that costs something and buys nothing.
The ceiling: never remove a correctness or security safeguard to save a call.
Validation, error handling, capped retries with backoff on genuinely flaky I/O, and audit
logging are not waste. Where this conflicts with pero-ponte-sueter, correctness wins.
Procedure
1. Name the cost unit and what it scales with. Requests, rows, tokens, milliseconds, megabytes, dollars, CI minutes, reviewer minutes — and per what: per call, per item, per user, per run. "Slow" and "expensive" are not units.
2. Measure it, or bound it in writing. One of these, before you change a line:
| Way | What you produce |
|---|---|
| Count | Actual call, query, or request count from a log, counter, or query log |
| Time | A profile or a timer around the suspected path, with the input that produced it |
| Bound | An analytic bound stated with the input size: "N orders → 1 + N queries; N is 430 in production" |
| Bill | The usage or billing dashboard figure for the endpoint, model, or job |
If you cannot measure it, write the bound and the input size that produces it. If you can do neither, say so and stop — you are guessing, and guessing is how the wrong line gets rewritten.
3. Find the waste. Read the measured path against these rows:
| Where it leaks | What it looks like |
|---|---|
| Data access | N+1 queries; refetching data already in hand; missing pagination, LIMIT, or result cap |
| Loops | A per-item call where one batch would serve; a value recomputed every iteration |
| Model and LLM calls | Whole file sent when one function would do; the largest model on a trivial task; unchanged context resent instead of cached; a failed prompt retried without changing it |
| Waiting | Polling where an event, webhook, or subscription exists; fixed sleeps longer than the work |
| Retries | Unbounded retries; no cap; retrying a deterministic failure that will fail identically |
| Resources | Connections, file handles, clients, or subscriptions opened and never closed; a client constructed per call |
| Build and CI | A full rebuild where incremental works; CI re-running suites the diff cannot affect; a cache configured but never restored |
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 · 96 lines · 77 tokens per session scan A 07c4278c4586
ni-que-fueramos-ricos is a skill published in the GitHub repository mcasillas17/mexican-mom (2 stars, last pushed 29d ago), licensed MIT. It adds 77 tokens to every session and 1,218 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
cco-tools
Show what tools actually cost in tokens — learned per-tool averages from observed results, replacing the hardcoded MCP/Agent guesses.
cco-clean
Clean up old tracking data and reset statistics.
stuck
Diagnoses a broken course setup by reading the folder, so the student never has to describe the problem in English. Takes no argument. Reports which class they are on and which files are filled, then names one blocker and one fix: Claude Code opened at the wrong level, .claude/skills/ missing, skills not loading…
magic-compact
Compact this Claude Code session.
sre-bot
Answer questions about production health and investigate incidents using live Kubernetes and observability data. Invoke whenever someone asks whether something is broken, slow, erroring, or down; asks what happened during an outage or time window; asks about alerts, logs, metrics, traces, or error rates; asks why a…
weather
Look up a location's weather forecast using a live web search. Invoke whenever the user asks about the weather, whether to expect rain, temperature, or what to wear or plan for outdoor activities.