CRMay 8

Demystifying and Detecting Agentic Workflow Injection Vulnerabilities in GitHub Actions

arXiv:2605.0713593.3
AI Analysis

For developers and security researchers using GitHub Actions with LLM-based agents, this paper identifies and quantifies a new class of injection vulnerabilities, with high precision and real-world impact (24 accepted/fixed cases).

This paper introduces Agentic Workflow Injection (AWI), a new vulnerability in GitHub Actions where untrusted event context is incorporated into LLM agent prompts or scripts, leading to attacker-influenced behavior. Using a taint-analysis tool (TaintAWI) on 13,392 workflows, they found 519 potential AWI vulnerabilities, with 496 confirmed exploitable (95.6% precision), including 343 zero-day vulnerabilities.

GitHub Actions is increasingly used to deploy LLM-based agents for repository-centric tasks such as issue triage, pull-request review, code modification, and release assistance. These agentic workflows extend traditional CI/CD automation with agentic capabilities but also create a new injection surface. In this paper, we introduce Agentic Workflow Injection (AWI), a workflow-level injection flaw where untrusted GitHub event context, such as issue bodies, pull-request descriptions, or comments, is incorporated into agent prompts or agent-consumed inputs and converted into attacker-influenced behavior through agent tools or downstream workflow logic. We identify two core AWI patterns: Prompt-to-Agent (P2A), where untrusted content reaches an agent prompt boundary, and Prompt-to-Script (P2S), where attacker influence propagates through model- or agent-derived outputs into later scripts. We present the first systematic study of AWI in GitHub Actions. We characterize 1,033 real-world AI-assisted actions and extract AWI-specific taint specifications, including prompt boundaries, derived outputs, agentic capabilities, and access-control interfaces. Based on these specifications, we design TaintAWI, a taint-analysis tool that tracks flows from untrusted event context to agent prompt inputs and security-sensitive workflow sinks. Applying TaintAWI to 13,392 real-world agentic workflows from 10,792 repositories, we report 519 potential AWI vulnerabilities, of which 496 are confirmed exploitable under our threat model, yielding a precision of 95.6%. Among them, 343 are previously unknown zero-day vulnerabilities. We prioritized disclosure for 187 zero-day cases, received 26 maintainer responses, and 24 cases have been accepted or fixed at the time of writing.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes