DANG-ai/SKILLER

Official implementation of SKILLER: Language-Level Reinforcement Learning for Reusable Skill Extraction in Small Language Models.

4Stars on the repository
71Mods indexed here, across every type
1mo agoLast push, which is what freshness is scored on
customA LICENSE file GitHub cannot name, so bodies are not copied

l2a3

50

DANG-ai/SKILLER

Skill Claude CodeCodex

After the last write, reread only the produced artifacts.

not rated 4 1mo ago A 0 tokens

u11-clean

51

DANG-ai/SKILLER

Skill Claude CodeCodex

Keep a small internal ledger, not a file, with one row per.

not rated 4 1mo ago A 0 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 28 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 26 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 26 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 29 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 26 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 28 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 24 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 22 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 26 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 28 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 27 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 24 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 28 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 27 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 24 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 27 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 23 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 24 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 25 tokens

DANG-ai/SKILLER

Skill Claude Code

Instance-memory boundary: distill only reusable procedures. Do not write concrete task IDs, sample IDs, training filenames or paths, final answers, option mappings, URLs, named entities, dates, numeric constants, or verbatim question text from the source trajectory into any generated SKILL.md. Use placeholders such as…

not rated 4 1mo ago A 28 tokens

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: