webpro255

awesome-ai-agent-attacks: Browse Sourced AI Agent Security Incidents

A curated, dated timeline of real AI agent security incidents, breaches, and vulnerabilities from 2024 to 2026. Use it to research documented attacks and their impacts, root causes, CVEs, and sources.

View on GitHub Compare
Stars
91
Forks
10
Stars, last 30 days
+9
Last commit
today
Contributors
2

Our take

Activeregular commits

Health score

27/100

How it is scored
Activity
35
Community
31
Issues
21
Pull requests
20

+9 stars in the last 30 days

  • Fewer contributors than 86% of the projects we track

This is a focused, actively updated reference list with useful sourcing, but its small, concentrated contributor base means readers should verify important claims against the linked sources.

Good fit if

  • You want a browsable timeline of sourced AI agent security incidents rather than a security tool.
  • You need incident summaries that call out impacts, root causes, and CVEs where applicable.
  • You are comfortable consulting the original sources for details and independent verification.

Look elsewhere if

  • You need software that detects, prevents, or responds to attacks.
  • You need a maintained database with releases or a documented issue-resolution process; the project has no releases and no recent issue or pull-request handling recorded.
  • You need a broad, independently maintained reference: most contributions come from one of its two contributors.
All health signals
Last commit2026-11-13 (today)
Commits, last 90 days13
Releases, last 12 months0
Contributors2 (top contributor: 71% of commits)
Issues closed, last 90 days0
Pull requests merged, last 90 days0
Project age7 months

Checked on 2026-11-14 with the GitHub API.

Project history

Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.

Selected snapshot

Stars

91

Forks

10

Change since Nov 17, 2026: 0 stars·0 forks

Overview

awesome-ai-agent-attacks is an Awesome-style reference list of real AI agent security incidents, breaches, and vulnerabilities from 2024 to 2026. Each entry is framed around a date, named target, specific impact, root cause, relevant CVEs, and source links, making it easier to check what happened rather than rely on unsourced claims.

The list is intended for security researchers, incident responders, and people assessing risks in agentic AI and related systems. The current quarter is on the main page, with earlier periods linked as archives. It is a research resource, not an attack-detection or mitigation tool.

Key Features

  • Dated incident entries covering 2024 through 2026.
  • Identifies the affected company, product, or target where reported.
  • Summarizes the concrete impact and stated root cause.
  • Includes CVE identifiers when applicable.
  • Links entries to external reporting and primary sources where available.
  • Separates earlier incidents into linked year and quarter archives.
  • Frames its contents as sourced facts rather than commentary or product recommendations.

Use Cases

  • Security researchers can trace public incidents involving AI agents and related infrastructure.
  • Incident responders can consult documented attack patterns and sources during threat research.
  • Security teams evaluating agent deployments can review examples of failures involving credentials, tool permissions, prompt injection, and supply chains.
  • Writers and educators can use the dated entries as a starting point for further research, following the cited sources for context.

License in plain words

MITpermissive

  • Commercial use: yes
  • Modify and redistribute: yes
  • You must keep: the copyright and license notice
  • Share your changes: no

A summary, not legal advice: the LICENSE file is what applies.

Getting Started

Browse the repository, then follow the links on its main page to the relevant incident period. Check the linked sources for details beyond each summary.

Considerations

The project is a curated list, so its coverage and summaries depend on what contributors include and on the quality and availability of cited sources. It is young and has no published releases; while commits have been regular recently, most of the work is concentrated with one contributor. Treat the entries as research leads and consult their sources, especially when using them to make security decisions.

Found this useful?

Share it with someone who would like awesome-ai-agent-attacks.