secret-scan-gitleaks
Fail
Audited by Gen Agent Trust Hub on Feb 14, 2026
Risk Level: HIGHCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (HIGH): The skill requires running the
gitleaksbinary on the host. This executes code against potentially untrusted directories, posing a risk of binary exploitation or unintended system impact if the tool is misconfigured or vulnerable. - [CREDENTIALS_UNSAFE] (HIGH): The skill's core function is to find and extract hardcoded secrets. This brings high-value credentials into the agent's session context, where they could be exposed or misused if the agent is subsequently compromised.
- [PROMPT_INJECTION] (HIGH): Indirect Prompt Injection vulnerability surface. The skill processes untrusted external content (code and git history) and outputs it for agent interpretation. (1) Ingestion points: Repository source code and git metadata via
gitleaks detectinSKILL.md. (2) Boundary markers: Absent. (3) Capability inventory: Command execution and local file read access. (4) Sanitization: Redaction of output is requested as a soft instruction but is not enforced by the tool chain itself.
Recommendations
- AI detected serious security threats
Audit Metadata