military-communicator

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill operates as a communication hub designed to ingest, process, and forward information between various roles ('Receive', 'Transfer', 'Confirm'), creating a vulnerability surface for indirect prompt injection where untrusted external data could influence agent behavior.\n
  • Ingestion points: Processes external 'communication tasks', 'information flow', and messages from various channels (SKILL.md, SKILL.v2.md).\n
  • Boundary markers: Lacks explicit delimiters or 'ignore' instructions for the data being handled, increasing the risk of the agent obeying instructions embedded in the processed information.\n
  • Capability inventory: The skill claims capabilities for technical support, tool maintenance, and permission management ('权限管理'), which could be exploited if malicious commands are injected into the message flow.\n
  • Sanitization: No sanitization, validation, or filtering mechanisms are specified for handling the transferred content.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 06:30 AM
Security Audit — agent-trust-hub — military-communicator