make-ai-safe-copy
Pass
Audited by Gen Agent Trust Hub on Aug 2, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a bundled Node.js script (
scripts/safectx.mjs) to perform local file sanitization. The execution is scoped to the skill's own scripts and acts on user-provided local file paths. - [DATA_EXFILTRATION]: The skill implements a 'privacy first' design. It explicitly instructs the agent not to read original files into the model context and to use only the sanitized versions. There are no network operations or external data transmissions detected.
- [PROMPT_INJECTION]: The instructions contain strict behavioral boundaries (e.g., 'Do not read... the original file contents into model context') designed to prevent accidental data exposure. These are safety instructions, not malicious overrides.
- [CREDENTIALS_UNSAFE]: While the script contains regular expressions for detecting AWS keys and GitHub tokens, these are used for redaction purposes. No hardcoded secrets or credentials were found.
- [EXTERNAL_DOWNLOADS]: No external dependencies or remote scripts are downloaded or executed. The skill uses standard Node.js built-in modules.
Audit Metadata