Agent skills have emerged as an effective means of enabling LLM agents to solve complex, interactive tasks by encoding reusable knowledge without modifying model parameters. Recent approaches further improve skill adaptation by distilling reusable experience from agent execution trajectories and evolving skills based on downstream task feedback. However, existing methods primarily represent skills as natural language documents, intertwining procedural logic with descriptive knowledge and making skills difficult to maintain and evolve. To address this limitation, we propose PASE (Program-Aided Patchable Skill Evolution), a program-aided framework that iteratively evolves skills by associating them with executable programs. Unlike prior approaches that encode trajectory-derived experience primarily in natural language, PASE separates reusable procedural logic into editable programs, enabling targeted updates to executable components while reducing the need to repeatedly revise natural language skills. Extensive experiments across diverse models and benchmarks demonstrate that PASE consistently outperforms all baselines, by achieving over 2.8% improvements. Further analyses show that PASE exhibits greater evolutionary stability and generalisation capability, highlighting the critical role of programmatic knowledge in skill evolution. We will release all data and code via GitHub.
Figures & tables
Figure 1: Comparison between vanilla text-only skill-based agents and the PASE framework.
Figure 2: Overview of our PASE framework.
Methods
In Domain
Out of Domain
Avg.↑
SpreadsheetBench
Domain-Spreadsheet
TableQA
Vrf↑
Soft↑
Hard↑
FIN↑
HR↑
RE↑
SAL↑
SC↑
Overall↑
WikiTQ↑
HiTab↑
Qwen3.6-27B
No Skill
43.81
19.57
20.76
26.17
18.02
23.97
10.34
24.13
21.07
51.35
42.63
32.04
Human Skill
64.76
42.67
25.58
53.09
41.80
43.45
38.70
42.36
44.02
70.23
46.90
48.97
LLM Skill
43.69
31.75
17.69
28.64
30.09
33.33
27.20
37.75
31.91
83.45
67.60
46.16
Table 1: Overall performance.
Methods
In Domain
Out of Domain
Avg.↑
SpreadsheetBench
Domain-Spreadsheet
TableQA
Vanilla LLM
43.81
21.07
42.63
35.84
w/ Text Skill
78.93
49.68
71.31
66.64
w/ Programmatic Skill
80.71
50.17
72.61
67.83
PASE
82.98
50.76
74.10
69.28
w/o Program Evolution
77.26
46.73
69.64
64.54
Table 2: Ablation study of PASE.
Figure 3: Skill evolution characteristics of PASE. Figure 3(a) analyzes the statistics of skill updates, while Figures 3(b) and 3(c) characterize the evolution and operation of the update mechanism. Figure 3(d) compares textual skill lengths, illustrating how executable programs help reduce the textual burden of skill evolution.
Figure 4: Generalization of skills learned by different skill evolution methods.
Appendix figures & tables5 assets
Supplementary material from the paper’s appendix.
Appendix
Dataset
# Examples
SpreadsheetBench
912
Verified
400
Full
912
Domain-Spreadsheet
677
FIN
135
HR
185
Appendix
Table 3: Statistics of the evaluation datasets.
Skill Creator
In Domain
Out of Domain
Avg.↑
SpreadsheetBench
Domain-Spreadsheet
TableQA
Vrf↑
Soft↑
Hard↑
FIN↑
HR↑
RE↑
SAL↑
SC↑
Overall↑
WikiTQ↑
HiTab↑
Skill User: Qwen3.6-27B
No Skill
43.81
19.57
20.76
26.17
18.02
23.97
10.34
24.13
21.07
51.35
42.63
32.04
Qwen3.6-27B
82.98
53.07
37.57
54.32
50.45
51.69
39.85
53.22
50.76
85.39
74.10
62.79
Gemma4-31B
76.90
51.66
35.75
62.47
52.79
49.81
41.38
60.41
54.90
85.31
77.76
63.74
Appendix
Table 4: Cross-model skill transfer results.
Figure 5: Spreadsheet Inference Agent Prompt.
Figure 6: Proposer Agent Prompt.
Case: Spreadsheet Manipulation: Extract and Look up
Instruction: For each string in cells A3:A14, extract the 2-digit month code and look up its corresponding month name in columns D:E. Fill cells B3:B14 with the results. Ground Truth Answer: B3 = SEP, B4 = JAN, B5 = DEC, …
Trace2Skill (Text-only Skill):
name: Month Code Lookup
description: Extract month codes from spreadsheet strings and map them to month names using a lookup table.
## When to Apply
Use when spreadsheet cells contain embedded month codes that need to be mapped to corresponding month names using a reference table.
Appendix
Table 5: Illustrative case study comparing text-only skill evolution with PASE on spreadsheet manipulation. Potentially incorrect knowledge and operations are highlighted in red , while correctly refined textual knowledge and program logic are highlighted in blue .