legal-tos-privacy
Warn
Audited by Gen Agent Trust Hub on Feb 23, 2026
Risk Level: MEDIUMDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill instructs the agent to access sensitive file paths and scan the codebase for credential-related strings, creating a risk of accidental exposure of secrets.\n
- Evidence: Instructions in
SKILL.mdto read.env.exampleand.env.local.examplefiles.\n - Evidence: Directives to search for patterns such as
password,secret,token,jwt,bearer, andauthorizationin source code.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its ingestion of untrusted data from the filesystem and external sites.\n - Ingestion points: The skill reads project metadata, READMEs, marketing content, and database schemas.\n
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the audit instructions.\n
- Capability inventory: The agent has the capability to perform comprehensive file reads and string searches within the project directory.\n
- Sanitization: No sanitization or validation is performed on the extracted data before it is incorporated into the drafted documents.
Audit Metadata