melech-pre-plan
Pass
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill uses behavioral framing to define a specific persona ("You are not plan mode"). These instructions are benign and intended to guide the agent's interaction style for design discussions rather than bypassing safety filters.
- [DATA_EXFILTRATION]: There are no commands for network communication (e.g., curl, wget) or access to sensitive local files (e.g., credentials, ssh keys). The skill operates entirely within the chat context.
- [REMOTE_CODE_EXECUTION]: The skill does not reference any external scripts, binaries, or remote code sources. It contains only natural language instructions.
- [COMMAND_EXECUTION]: No shell commands or subprocess calls are present. The frontmatter includes 'disable-model-invocation: true', which restricts the agent's ability to trigger external tools or models.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes user-provided task descriptions, it does so to facilitate a dialogue. The surface for injection is minimal as it does not execute code or perform high-privilege operations based on that input.
Audit Metadata