invent-by-hand
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill consists entirely of markdown instructions and templates. No scripts, binaries, or command-line execution patterns are present. The methodology explicitly mandates "by hand, no code" for its output.
- [EXTERNAL_DOWNLOADS]: There are no remote resources, external URLs, or package manager commands (npm, pip, etc.) within the skill files.
- [DATA_EXFILTRATION]: No network-capable commands (curl, wget) or file system operations accessing sensitive paths were identified. The skill operates purely on text input provided by the user during the interaction.
- [PROMPT_INJECTION]: No malicious instruction overrides or jailbreak attempts were detected. The skill includes adversarial test cases designed to enforce logical constraints, such as refusing to invent concepts for vague inputs, which enhances agent reliability.
- [OBFUSCATION]: The content is fully transparent. No Base64 encoding, zero-width characters, homoglyphs, or hidden text patterns were found in the instructions, examples, or metadata.
Audit Metadata