scout
Fail
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: HIGHDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill is instructed to access and read sensitive credential files to facilitate API interactions.
- Evidence: The instructions direct the agent to retrieve YouTube API keys from
~/.config/jack-keys.env. - Exfiltration Risk: While the data is used for legitimate API calls, the agent also has the capability to send information to external channels (Telegram), creating a potential path for credential exposure.
- [PROMPT_INJECTION]: The skill processes untrusted data from multiple external platforms, which could contain malicious instructions intended to influence the agent's behavior.
- Ingestion points: The skill reads YouTube comments, GitHub repository descriptions/trends, and social signals from X and Reddit via
agent-reach. - Boundary markers: There are no specific delimiters or instructions provided to the agent to distinguish between data and commands within the retrieved content.
- Capability inventory: The agent is authorized to write files to the user's desktop (
~/Desktop/📋 Notes & Markdown/) and transmit summaries over the network via Telegram. - Sanitization: The skill lacks explicit sanitization or validation steps for the content pulled from these external APIs.
Recommendations
- AI detected serious security threats
Audit Metadata