inbox
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION] (LOW): The skill is vulnerable to indirect prompt injection because it processes and summarizes external message content from the Agent Mail system without using delimiters or instructions to ignore embedded prompts.
- Ingestion points: Messages are retrieved via the
fetch_inboxandsearch_messagestools or an HTTP fallback tolocalhost:8765. - Boundary markers: None. The skill lacks explicit markers or safety instructions when presenting message data to the LLM.
- Capability inventory: The skill uses
curl,git, andhostnamefor coordination. - Sanitization: None. Message data is used directly for summaries and display.
- [COMMAND_EXECUTION] (SAFE): The skill executes local shell commands (
git,hostname,curl) to identify the agent and interact with a local service. These operations are appropriate for the skill's context and do not involve remote code execution or privilege escalation.
Audit Metadata