by 0x4D31 · Agent Tool · ★ 48
Endpoint AI Agent Abuse (EAA) A curated, evidence-scoped catalog of techniques for abusing endpoint-resident AI agents. EAA tracks how attackers can misuse local agents—especially coding and developer agents—as execution, persistence, collection, exfiltration, defense-evasion, and inherited-authority surfaces. The focus is the endpoint behavior around agents: launchers and flags, environment and provider routing, instructions, memory, hooks, plugins, skills, MCP, transcripts, telemetry, and access inherited from the local user.
| Stars | 48 |
| Forks | 2 |
| Language | Python |
| Category | Agent Tool |
| License | CC0-1.0 |
| Quality Score | 62.5825051506002/100 |
| Last Updated | 2026-08-16 |
| Created | 2026-06-26 |
| Platforms | python |
| Est. Tokens | ~15k |
Explore other popular agent tool tools:
endpoint-ai-agent-abuse is EAA is a curated catalog of techniques and real-world cases involving abuse of local AI agents through their runtime, configuration, state, tools, and inherited authority.. It is categorized as a Agent Tool with 48 GitHub stars.
endpoint-ai-agent-abuse is primarily written in Python.
You can find installation instructions and usage details in the endpoint-ai-agent-abuse GitHub repository at github.com/0x4D31/endpoint-ai-agent-abuse. The project has 48 stars and 2 forks, indicating an active community.
endpoint-ai-agent-abuse is released under the CC0-1.0 license, making it free to use and modify according to the license terms.