app-knowledge
Fail
Audited by Gen Agent Trust Hub on Mar 27, 2026
Risk Level: HIGHCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill documentation explicitly lists the
{base folder}/.envfile as a source for the agent to retrieve credentials. Environment files frequently contain plaintext sensitive data such as API keys, database connection strings, and authentication tokens. Accessing these files increases the risk of credential leakage or misuse. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface. 1. Ingestion points: Ingests data from codebase files, documentation folders, and environment files (referenced in SKILL.md). 2. Boundary markers: Absent; no instructions are provided to the agent to ignore or delimit embedded instructions in the files it reads. 3. Capability inventory: Uses browser-use for web automation and various codebase search/read tools (referenced in SKILL.md). 4. Sanitization: Absent; no validation or escaping of external file content is described.
Recommendations
- AI detected serious security threats
Audit Metadata