Prompt Warrior: a skill that turns Claude Code logs into an RPG character sheet
A developer has released Prompt Warrior, an open-source skill that turns Claude Code session logs into a game-style character profile: a title, a level, 74 achievements with Steam-style rarities, and a biography written by the AI itself. In a month and a half, the author had piled up about 200,000 words of these logs.
AI-processed from Habr AI; edited by Hamidun News
A developer has released Prompt Warrior, an open-source skill that turns local session logs from Claude Code and other AI harnesses into a gamified psychological portrait of the user, complete with a title, a level, 74 achievements, and a biography written by the AI itself. The project is on GitHub as timoncool/prompt-warrior.
How the idea came about
The author says he has been living in Claude Code for several months now: assigning tasks, cursing, asking for redos, and cursing again. At some point he realized that every line he typed was quietly being saved to disk. Every "do it," every "check it," every all-caps message at 3 a.m. was piling up in JSONL files in his home directory. In a month and a half, he had accumulated about 200,000 words of these logs, roughly the length of two copies of The Master and Margarita. That realization — how much text had built up without him noticing — became the starting point for the project.
How the RPG character sheet works
Looking at that mountain of data, the author wondered how he looked from the machine's side: which words he used to give orders, what hour of the day he was most "dangerous," how many times he cut the model off mid-sentence. Over a single day of conversation with the AI, he built Prompt Warrior, a skill that analyzes the logs and assigns the user a title, a level, and a profile built on six character scales.
The concept of turning work logs into a game profile borrows familiar mechanics from video games and Steam: achievement rarities, stat scales, titles, and levels make dry statistics about commands and sessions emotionally legible. Instead of a table showing how many prompts you sent this month, you get a character with a backstory — and the AI writes that biography itself, drawing on the same pool of logs that feeds all the other stats.
- The project is called Prompt Warrior; the repository is github.com/timoncool/prompt-warrior
- In a month and a half, the author accumulated about 200,000 words of Claude Code logs
- The profile includes 74 achievements with Steam-style rarities
- Character assessment uses six scales, and the AI writes the biography itself
What it means
Prompt Warrior turns routine logs from AI tools into a game mechanic for self-reflection: instead of dry usage stats, the user gets a character with a biography of their own, and a clear look at how much text their everyday conversations with the model actually produce. The project also shows how much data about working style and behavior builds up in ordinary session logs from Claude Code and other harnesses, and how useful that data can be for analysis beyond its original purpose. As an open-source skill, Prompt Warrior is available to anyone using Claude Code or other supported harnesses who has accumulated local session logs of their own.
FAQ
Where can I get Prompt Warrior?
The skill is publicly available on GitHub in the timoncool/prompt-warrior repository. It works with Claude Code and other AI harnesses.
How many logs did the project's author accumulate?
In a month and a half of using Claude Code, the author accumulated about 200,000 words in JSONL logs, roughly the length of two copies of The Master and Margarita.
- Nov 1, 2026Local models lose to Opus and Codex: 0.21 on a MacBook versus 1.00 for Codex
- Nov 1, 2026Kimi scores 96.2% on a 1C database analytics test and emerges as a fallback if Claude access is blocked
- Nov 1, 2026OpenAI Rewrites GPT Behavior Rules: Five Values and a Chain of Command
- Nov 1, 2026Google admits AI Mode cuts search revenue, shares fall 15%
Key Questions Answered
Where can I get Prompt Warrior?
The skill is publicly available on GitHub in the timoncool/prompt-warrior repository. It works with Claude Code and other AI harnesses.
How many logs did the project's author accumulate?
In a month and a half of using Claude Code, the author accumulated about 200,000 words in JSONL logs, roughly the length of two copies of The Master and Margarita.
How the idea came about?
The author says he has been living in Claude Code for several months now: assigning tasks, cursing, asking for redos, and cursing again. At some point he realized that every line he typed was quietly being saved to disk. Every "do it," every "check it," every all-caps message at 3 a.m. was piling up in JSONL files in his home directory. In a month and a half, he had accumulated about 200,000 words of these logs, roughly the length of two copies of The Master and Margarita. That realization — how much text had built up without him noticing — became the starting point for the
Build & Ship Production AI Agents Without Boilerplate Code
Start your free 72-hour trial at Hamidun Academy (academy.hamidun.com) — interactive lessons on Claude Code & Cursor, live Tech Radar, and production agent templates.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.