Explore/framework/AgentPoison
A

AI-secure/AgentPoisonActive

[NeurIPS 2024] Official implementation for "AgentPoison: Red-teaming LLM Agents via Memory or Knowledge Base Backdoor Poisoning"

frameworkPythonMIT
GitHubCompare
Refreshed 5h ago
OverviewActivity52wAlternativesDocs
Stars238
Forks33
HF Downloads30d
Last commit9d ago
Refreshed5h ago
Project healthActiveLast commit 9d ago.
Production readinessExperimentalGrowing but not yet battle-tested at scale.
Risk notesMITVerify license before production use.
AgentHub Score
68 / 100
Composite score from 6 signals. How we score →
Active project
68Score
Growth
83B+
Activity
90A
Documentation
70C+
Maturity
71C+
Community
82B+
Production
58C
GitHub stars · 29 days observed238 not enough history
snapshots
Repository activity · 29 days observedReal snapshots from pushed_at
inactivepushed
2026-07-262026-08-25
Practical assessment
Should you use it?

✓ Best for

  • Multi-agent orchestration
  • Production agentic workflows
  • Stateful long-running tasks

◎ Strengths

  • Stable API
  • Active release cadence
  • Strong GitHub community

✕ Not ideal for

  • Simple single-step automation
  • Teams without Python/ML expertise

⚠ Watch-outs

  • Breaking changes between minor versions
  • Ecosystem lock-in if tightly coupled
Technical details
What's inside
LanguagePython
LicenseMIT
Sourcegithub
Open source✗ No
Commercial use
Docs
Demo
Paper

AgentHub Score

68
Score 68/100
Above average

Alternatives

E
ECC
243.2k · framework
95
H
hermes-agent
236.4k · framework
95
D
deepseek-harness
195.1k · framework
95
L
langchain
145.0k · framework
95
Compare all →

Recent activity

Latest commit 9d ago9d ago
Indexed by AgentHub crawler5h ago
Monitor for new releasesongoing