A minimal demonstration of the python preprocessor step. The phase
declares a Python function that runs in pure mode (sandboxed), computes
deterministic text statistics, and injects them into the artifact under
data.stats. The LLM then writes a commentary that references those
exact numbers — something LLMs are otherwise unreliable at.
Input
user_message text — any string.
Output
text_review.commentary — short prose discussing the input through the
lens of the precomputed statistics.
Why this is a good fit for python
LLMs are bad at counting characters / tokens / lines accurately. Python
counts them precisely in microseconds. By doing the count in Python and
showing the result to the LLM, the commentary stays factual.
1---2name: tya5-reyn-word-stats-demo3description: Overview4---56## Overview78A minimal demonstration of the `python` preprocessor step. The phase9declares a Python function that runs in pure mode (sandboxed), computes10deterministic text statistics, and injects them into the artifact under11`data.stats`. The LLM then writes a commentary that references those12exact numbers — something LLMs are otherwise unreliable at.1314## Input1516`user_message` text — any string.1718## Output1920`text_review.commentary` — short prose discussing the input through the21lens of the precomputed statistics.2223## Why this is a good fit for python2425LLMs are bad at counting characters / tokens / lines accurately. Python26counts them precisely in microseconds. By doing the count in Python and27showing the result to the LLM, the commentary stays factual.2829---30> Source: [tya5/reyn](https://github.com/tya5/reyn) — distributed by [TomeVault](https://tomevault.io).31<!-- tomevault:4.0:skill_md:2026-05-22 -->
Run npx skillmds@latest add tomevault-io/tya5-reyn-word-stats-demo in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.
Overview It is listed under Docs & Writing on SkillMD.
This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.
This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.
Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.
tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.
tomevault-io