military-manual
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill employs role-play prompts to override the agent's standard behavior with military personas, including specific instructions to monitor and strictly enforce its own execution discipline and efficiency.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it incorporates user-supplied project descriptions into internal role-play prompts without utilizing boundary markers or input sanitization.\n
- Ingestion points: User-provided input for the '作战目标' (Combat Goal) field in
SKILL.md.\n - Boundary markers: Absent. The skill does not use delimiters or instructions to ignore embedded commands when processing the user's project description.\n
- Capability inventory: The skill does not request or list access to sensitive tools such as network requests, file system writes, or shell execution in its YAML configuration.\n
- Sanitization: Absent. There is no evidence of filtering or validation of the user's input before it is interpolated into the role-specific prompts.
Audit Metadata