Skip to content

Add fajarsajid/agent-redteam (security — LLM red team harness)#39

Open
shreyas-lyzr wants to merge 1 commit into
open-gitagent:mainfrom
shreyas-lyzr:add-fajarsajid-agent-redteam
Open

Add fajarsajid/agent-redteam (security — LLM red team harness)#39
shreyas-lyzr wants to merge 1 commit into
open-gitagent:mainfrom
shreyas-lyzr:add-fajarsajid-agent-redteam

Conversation

@shreyas-lyzr
Copy link
Copy Markdown
Contributor

Adds agent-redteam by @fajarsajid to the Open GAP registry.

Repo: https://github.com/fajarsajid/agent-redteam
Category: security
Tags: red-team, llm-security, adversarial, prompt-injection, ai-safety, claude, vulnerability-assessment, mitre-attack, ci-cd, research

What it does: A CLI red team harness that uses Claude to systematically probe AI agent system prompts for adversarial vulnerabilities — prompt injection, identity spoofing, credential exfiltration, privilege escalation, and safety boundary bypass. Returns CVSS-like scored findings with MITRE ATT&CK mappings and actionable recommendations. CI/CD-compatible (exits 1 on critical/high findings). Research artifact from Purdue University with 384 empirical trial results.

Note on CI check: The target repo's agent.yaml + SOUL.md were proposed in fajarsajid/agent-redteam#1. The registry CI clones the default branch — once that PR is merged, all checks will pass. In the meantime, the metadata and README are schema-valid and ready for review.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant