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.
git clone --depth 1 https://github.com/UniM0cha/self-improving-skillsWrote 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/commands/unim0cha/self-improving-skills/distill-skill)<a href="https://agentmods.dev/commands/unim0cha/self-improving-skills/distill-skill"><img src="https://agentmods.dev/badge/commands/unim0cha/self-improving-skills/distill-skill/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/commands/unim0cha/self-improving-skills/distill-skill"><img src="https://agentmods.dev/badge/commands/unim0cha/self-improving-skills/distill-skill.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 | $0.00032 | $0.00592 |
| Opus 5.5 | $0.00013 | $0.00237 |
| Sonnet 5.5 | $0.00006 | $0.00118 |
| Haiku 4.5 | $0.00003 | $0.00059 |
Grade A, and why
distill-skill 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 today.
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.
What it actually says
방금 진행한 작업을 돌아보고, 재사용 가능한 기법·패턴·해결책이 있으면 학습 스킬로 증류하세요.
이것은 자기개선 루프의 수동 트리거입니다 (Stop 훅 자동 nudge와 동일한 일을 사용자가 원할 때 실행).
claude-code-self-improving-skills:skill-distiller 서브에이전트에 위임하세요:
Task(또는 Agent) 도구로 subagent_type="claude-code-self-improving-skills:skill-distiller" 를 run_in_background=true 로 호출하고 (플러그인 네임스페이스 접두사를 포함한 정확한 이름입니다 — skill-distiller 만으로는 호출이 실패합니다; 백그라운드로 돌려야 증류가 도는 동안 사용자 입력이 막히지 않습니다), 방금 세션에서 무엇을 했고 무엇이 비자명했는지(어떤 문제를 어떻게 풀었는지, 어떤 API/라이브러리 동작을 발견했는지)를 간단히 전달하세요. 서브에이전트가:
~/.claude/skills/**/SKILL.md를 훑어 관련된 기존 스킬이 있는지 확인하고,- 있으면 그 스킬을 patch, 없으면(그리고 class-level로 재사용 가능할 때만) 새 스킬 생성,
- 일회성·환경의존적·비재사용 지식이면 아무것도 만들지 않고 그 이유를 한 줄로 보고
합니다.
$ARGUMENTS 가 주어지면, 그 안내(예: 특정 기법에 집중, 또는 특정 기존 스킬을 갱신)를 서브에이전트에 함께 전달하세요.
이 플러그인의 증류는 Opus 이하 모델로만 돕니다(에이전트 frontmatter 가 model: opus). 환경변수 SIS_DISTILLER_MODEL 이 설정되어 있으면(예: sonnet) 호출에 model="<값>" 파라미터를 포함하세요 — 호출 시 지정한 model 이 frontmatter 보다 우선합니다. 값이 fable·mythos 처럼 Opus 위 티어면 opus 로 넘기세요.
참고: 증류는 "재사용 가능한 기법"이 있을 때만 의미가 있습니다. 단순 기능 구현·리팩토링·일회성 수정만 했다면 서브에이전트가 스스로 "만들 것 없음"으로 판단하고 종료하는 것이 정상입니다.
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.
- today First seen · 24 lines · 32 tokens per session scan A cfaf029cd045
distill-skill is a command published in the GitHub repository UniM0cha/self-improving-skills (51 stars, last pushed 1mo ago), licensed MIT. It adds 32 tokens to every session and 592 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-11-07.
distill-skill compared with commands from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| distill-skillUniM0cha/self-improving-skills | Command | 51 | 1mo ago | MIT | A | 32 | Claude Code |
| reportrohitg00/awesome-claude-code-toolkit | Command | 2.8k | 5mo ago | Apache-2.0 | A | 17 | Claude Code |
| env-setuprohitg00/awesome-claude-code-toolkit | Command | 2.8k | 5mo ago | Apache-2.0 | A | 0 | Claude Code |
| debug-podrohitg00/awesome-claude-code-toolkit | Command | 2.8k | 5mo ago | Apache-2.0 | A | 0 | Claude Code |
| document-allrohitg00/awesome-claude-code-toolkit | Command | 2.8k | 5mo ago | Apache-2.0 | A | 0 | Claude Code |
| fix-commentsrohitg00/awesome-claude-code-toolkit | Command | 2.8k | 5mo ago | Apache-2.0 | A | 13 | Claude Code |
| maprohitg00/awesome-claude-code-toolkit | Command | 2.8k | 5mo ago | Apache-2.0 | A | 0 | Claude Code |