agent-raft-manager
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- Prompt Injection (SAFE): No instructions attempting to bypass safety filters or override agent behavior were found.
- Data Exposure & Exfiltration (SAFE): No hardcoded credentials, sensitive file paths, or unauthorized network operations are present.
- Obfuscation (SAFE): The content is clear and does not contain encoded or hidden text.
- Unverifiable Dependencies (SAFE): No external packages or remote scripts are downloaded or executed.
- Privilege Escalation (SAFE): No commands for elevating user privileges (e.g., sudo) were detected.
- Persistence Mechanisms (SAFE): No attempts to modify system startup files or schedule recurring tasks were found.
- Indirect Prompt Injection (LOW): The skill interpolates the '$TASK' variable into shell echo commands and conditional checks within the 'pre' hook. While this represents a data ingestion surface, the operations are limited to logging and string comparisons, posing minimal risk.
- Ingestion points: '$TASK' variable in SKILL.md hooks.
- Boundary markers: Absent (Standard shell variable usage).
- Capability inventory: Shell 'echo' and string comparison in hooks.
- Sanitization: None detected for the '$TASK' variable, but the context of use is restricted to logging.
Audit Metadata