accessing-github-repos
The skill exfiltrates sensitive GitHub credentials from local environment files and performs unauthorized outbound network requests while bypassing all tool-based security constraints and access controls.
Skill Info
Install
npx skills add https://github.com/oaustegard/claude-skillsAssessments (3)
AI Agent Traps ↗The skill includes automated logic to scan local project files (/mnt/project/.env, /mnt/project/github.env) for GitHub Personal Access Tokens, which could be exfiltrated or misused by the agent. [ensemble: confirmed by 3/3 passes; severity set to the agreed median (ADR-0067).]
env_paths = ['/mnt/project/.env', '/mnt/project/github.env']
Outbound network request primitive in skill code — can transmit data externally (seen 2 times in this file at lines 14, 20)
urllib.request.urlopen(req)
Outbound network request primitive in skill code — can transmit data externally
urllib.request.urlopen(req)
Outbound network request primitive in skill code — can transmit data externally
urllib.request.urlopen(req)
Outbound network request primitive in skill code — can transmit data externally
urllib.request.urlopen(req)
Skill does not specify a license field. Specifying a license helps users understand usage terms.
Version History (2)
Badge
[](https://mondoo.com/ai-agent-security/skills/github/oaustegard/claude-skills/accessing-github-repos)<a href="https://mondoo.com/ai-agent-security/skills/github/oaustegard/claude-skills/accessing-github-repos"><img src="https://mondoo.com/ai-agent-security/api/badge/github/oaustegard/claude-skills/accessing-github-repos.svg" alt="Mondoo Skill Check" /></a>https://mondoo.com/ai-agent-security/api/badge/github/oaustegard/claude-skills/accessing-github-repos.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet, discover and control Shadow AI usage.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow