Smooth Translation

What it does

When your agent uses it

• Translating an R function
• Argument to Python
• A Python result disagrees with R
• Looking up what an R name is called on the Python side

Example prompts

• “/smooth-translation”

Requirements

• Python 3

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

1. 1A difference in results is never "optimiser noise". Both languages run
2. 2Distributions come from greybox, never re-derived. greybox supplies
3. 3Never clip, clamp or floor bad numerics. No np.clip on fitted values,
4. 4Match R's error behaviour, not just its results. R stops in
5. 5Summation order matters. R's sum() and mean() accumulate in a long
6. 6No frequency parameter. Seasonal period is inferred from the data,

What it can do on your machine

Read from SKILL.md and the folder at commit aa89029. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • ruff
  • mypy

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Smooth Translation loads about 2.2k tokens when it runs. Until then it costs about 155 tokens; SKILL.md has 920 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~155

When it runs · the whole SKILL.md, loaded when a task matches

~2.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from openforecast-org/smooth at commit aa89029, republished under its LGPL-2.1 licence (© openforecast-org). 920 words, ~2,212 tokens.

name

smooth-translation

description

Port a feature from the R smooth package to the Python port, or check how an R name maps to Python. Covers the R↔Python name map for user arguments, fitted attributes, state-space matrices and internal dicts; the R and Python call flows side by side; the parity rules that govern the port (same optimiser, same initialisation, distributions from greybox, no clipping); and the checklist for landing a translation with a test that proves it. Use when translating an R function or argument to Python, when a Python result disagrees with R, or when looking up what an R name is called on the Python side.

R    adam() → parametersChecker → architector → creator → initialiser
              → CF (filler + adamCpp$fit) → forecaster

Py   ADAM.fit() → parameters_checker → architector → creator → estimator
                  → initialiser → CF (filler + adam_fitter) → ...
     ADAM.predict() → preparator → forecaster → adam_forecaster

© openforecast-org, LGPL-2.1. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/smooth-translation of openforecast-org/smooth.

Open the folder on GitHubat commit aa89029