{"name":"skill-distill","repo":"sybil-solutions/skill-distill","url":"https://midorreal.com/project/sybil-solutions-skill-distill","repository":"https://github.com/sybil-solutions/skill-distill","description":"Distill a good Claude Code / Codex session into a skill an open-weight model can follow to reproduce the task","facts":{"license":{"value":"MIT","source":"https://github.com/sybil-solutions/skill-distill","checked":"22 Nov 2026"},"language":{"value":"Python","source":"https://github.com/sybil-solutions/skill-distill","checked":"22 Nov 2026"},"releases":{"value":"0","source":"https://github.com/sybil-solutions/skill-distill/releases","checked":"21 Nov 2026"},"contributors":{"value":"1","source":"https://github.com/sybil-solutions/skill-distill/graphs/contributors","checked":"21 Nov 2026"},"archived":{"value":"no","source":"https://github.com/sybil-solutions/skill-distill","checked":"22 Nov 2026"}},"stars":405.0,"writeup":{"what_it_is":"skill-distill is a skill for Claude Code or Codex that turns a good session into a skill an open-weight model can follow. The teacher writes the skill, runs a student harness and model in tmux on a clean copy of the starting state, compares results with the reference, and revises the skill until the student matches.","audience":"Developers who want open-weight models to reproduce tasks done by frontier models","claims":[{"kind":"specific","claim":"Runs three trials per skill version","excerpt":"Three trials per skill version. One lucky run proves nothing.","status":"not_checked"},{"kind":"specific","claim":"Provides an adapter for omp (oh-my-pi) as a tested student harness","excerpt":"| omp (oh-my-pi) | tested |","status":"not_checked"},{"kind":"specific","claim":"Converts Claude Code, Codex and omp session logs into readable references","excerpt":"Claude Code / Codex / omp session log → readable reference","status":"not_checked"},{"kind":"specific","claim":"Requires a final human acceptance before a skill is considered done","excerpt":"the skill counts as done only when you accept it.","status":"not_checked"}],"alternatives":[],"written_by":"AI, from the project's README","date":"2026-11-22"},"why_now":null,"owner_supplied":null,"score":{"verdict":"mid","hype":42,"reality":29,"gap":13,"momentum":null,"confidence":60,"version":"score-2.0","rule":"verdict-2.0","calculated_at":"2026-11-22T07:00:00+00:00","verdict_source":"rule"},"early_read":null,"changes":[{"date":"2026-11-21T01:43:00+00:00","text":"Repository created","source":"https://github.com/sybil-solutions/skill-distill"}],"cite":"skill-distill is a GitHub project at https://github.com/sybil-solutions/skill-distill written mainly in Python, described there as \"Distill a good Claude Code / Codex session into a skill an open-weight model can follow to reproduce the task\".","attribution":"Data from Mid or Real, https://midorreal.com/project/sybil-solutions-skill-distill, checked 22 November 2026. Attribution with a link is required.","still_real":null}