Paper page - Grounded Skill Synthesis from Code at Scale for Agentic Intelligence

Source: https://huggingface.co/papers/2609.05571

Abstract

Code2Skill automatically extracts verifiable procedural skills from source code to improve agent performance before interaction experience accumulates.

Reusable skills give agents transferable procedural knowledge, making scalable acquisition essential for extending agents beyond prior experience. Existing methods face two limitations: trajectory-based synthesis requires interactions with specific environments, while document-derived skills may lack executable evidence and verification. Source code offers a complementary path: it requires no prior agent experience yet provides executable evidence for grounding abstractions. We presentCode2Skill, a fully automated pipeline that transforms selected code units into implementation-anchored records of atomic operations, composite workflows, and recurring patterns, then verifies each record throughsource-body-blind reconstructionandsource-aware comparison. Applied to 19,769 popular, actively maintained GitHub repositories,Code2SkillproducesCodeSkillBank, a grounded bank of 1,006,822 accepted records with workflow, boundary, provenance, and source-evidence metadata. Across 72 protocol-matched evaluations covering nine model settings and eight benchmarks, models augmented with retrievedCodeSkillBankskills improve by 11.7% on average over matched baselines and outperform them in 57 cases. Under a unified downstream interface,Code2Skillalso outperformstrajectory-derived skill bankson all seven shared benchmarks, showing that repository-derived skills can provide useful procedural knowledge before agents accumulate sufficient interaction experience. Skills synthesized from testedAI-generated codeachieve a 93.50% pass rate, compared with 93.00% for human-written code, providing initial evidence that the pipeline can expand with the growing volume of AI-generated software. Overall,Code2Skilltransforms procedural knowledge embedded in repositories into grounded, verifiable, and transferable agent skills.

View arXiv pageView PDFProject pageGitHub10Add to collection

Get this paper in your agent:

hf papers read 2609\.05571

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.05571 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.05571 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.05571 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.