Intel / article

Specifying and Maintaining Agentic Workflows: An Empirical Study of GitHub Agentic Workflows

Source: arxiv.org
Author: Jasem Khelifi, Issam Oukhay, Ali Ouni, Mohammed Sayagh, Mohamed Aymen Saied
Date: Published Sep 24, 2026

Context

GitHub Agentic Workflows turns written instructions and configuration into recurring automated work. An earlier documentation example shows the purpose: an drafts documentation changes for an engineer to review, as detailed in GitHub's account.

A study of 1,248 workflow files across 276 repositories finds that these instructions continue evolving. Among files observed for at least 120 days, 78.2% still received updates in the fourth month. The study therefore treats written agent instructions as maintained operational specifications. Teams copying a workflow inherit instructions that may need revision as the recurring task and its constraints change.

Key takeaways · AI-distilled

gh-aw workflow instructions are long specifications rather than short prompts: the median file runs 556.5 words, and 62.1% include code blocks.
The files keep changing: 78.2% of those observed for at least 120 days were still being updated in month four, though size-normalized churn drops after the first month.
Tasks, outputs, constraints and process instructions each appear in over 93% of labeled workflows, yet only 9.4% explicitly address prompt-injection defense.
The authors recommend tracking how workflows copied across repositories evolve and adding prompt-injection defenses, resource budgets and evidence-credibility checks where applicable.

Terms in this piece · Glossary

AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.

Read the source arxiv.org