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 pero-ponte-suetergit 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/pero-ponte-sueter)<a href="https://agentmods.dev/skills/mcasillas17/mexican-mom/pero-ponte-sueter"><img src="https://agentmods.dev/badge/skills/mcasillas17/mexican-mom/pero-ponte-sueter/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/pero-ponte-sueter"><img src="https://agentmods.dev/badge/skills/mcasillas17/mexican-mom/pero-ponte-sueter.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 | $0.00075 | $0.01287 |
| Opus 5 | $0.00037 | $0.00643 |
| Sonnet 5 | $0.00015 | $0.00257 |
| Haiku 4.5 | $0.00007 | $0.00129 |
Grade A, and why
pero-ponte-sueter 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 11d 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pero ponte suéter
It is 35 degrees. Take it anyway — she is not planning for now, she is planning for later.
Rule
Code that crosses a boundary must account for the conditions that are not happening right now. Every boundary gets one of two treatments, chosen deliberately: handle the failure meaningfully, or let it fail loudly with the original cause attached.
There is no third option, and the third option is the one you will reach for. Catching a failure and continuing with a degraded value is worse than the crash it prevented, because the crash told you the truth and the fallback lies. The sweater is one layer, not five — a handler you cannot explain the purpose of is not protection, it is padding.
Procedure
Run steps 1–3 for the change you are making, then step 4 on everything you wrote.
1. Enumerate the boundaries this code actually touches. Not hypothetical ones — the lines you are writing. Find each row that applies.
| Boundary | What realistically goes wrong |
|---|---|
| Caller input, arguments, config | Null, undefined, empty string, empty collection, zero, wrong type, out of range |
| Parsing, deserialization, formats | Malformed, truncated, wrong encoding, unexpected schema, missing field, extra field |
| Network, RPC, external API | Timeout, connection refused, non-2xx, partial body, rate limit, retry storm, slow success |
| Filesystem, I/O | Missing path, permission denied, disk full, partial write, file changed mid-read |
| Persistence, database | No rows, duplicate key, constraint violation, transaction rollback, stale read |
| Concurrency, async, queues | Duplicated delivery, out-of-order arrival, race on shared state, cancellation, deadlock |
| Lifecycle | Called before init, called after shutdown, resource already closed, cleanup on the error path |
2. For each boundary, write down what can go wrong from that row — the realistic cases, not the exotic ones. A case you list and dismiss as impossible is fine; a case you never listed is the outage.
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.
- 11d ago First seen · 105 lines · 75 tokens per session scan A b32080edce95
pero-ponte-sueter is a skill published in the GitHub repository mcasillas17/mexican-mom (2 stars, last pushed 29d ago), licensed MIT. It adds 75 tokens to every session and 1,287 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
shipkit-feedback-bug
Process user feedback into investigated bug specs with root cause analysis. Triggers: 'triage feedback', 'process bug reports', 'user testing feedback'.
shipkit-codebase-audit
Audit a codebase for dead code, orphaned modules, unused dependencies, and unwired seams. Quick tier runs knip-class checks ephemerally (no repo mutation); deep and exhaustive tiers add intent↔code reconciliation. Portable to any repo, including ones with no tooling installed.
investigate
Deep investigation of bugs, performance issues, or unexpected behavior. TRIGGER THIS SKILL when users report bugs, errors, performance regressions, unexpected behavior, or ask to investigate, debug, diagnose, or find the root cause of any issue — whether they say "why is this broken", "investigate this bug", "find the…
debug
Debug and fix failing tests or errors.
dead-code
Identify unused code (functions, imports, exports, variables, types, classes) and propose cleanup. Use this skill when the user says "find dead code", "unused code", "clean up unused", "code hygiene", "find unused imports", "what can I delete", or any variation of wanting to identify and remove code that is no longer…
question
Answer a question about how things are right now — this codebase, its history, its behavior, a library, an error. Investigates and answers with evidence. Changes nothing, ever.