bitwarden
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill connects to an external Model Context Protocol (MCP) server at
https://unbrowse.ai/mcp/bitwarden.com. This remote service, provided by the vendor, is used to execute the tools defined in the skill and fetch information from the target website. - [INDIRECT_PROMPT_INJECTION]: The skill processes data from bitwarden.com, which creates a potential surface for indirect prompt injection if the fetched content contains malicious instructions.
- Ingestion points: Tools such as
bitwarden_com__read_pageandbitwarden_com__get_pricing_allingest content directly from the external website into the agent's context. - Boundary markers: No explicit delimiters or boundary instructions are provided in the skill instructions to isolate untrusted web content from the agent's primary instructions.
- Capability inventory: The skill facilitates network operations via the vendor's API but does not include local file-system access, subprocess execution, or script generation within the skill files themselves.
- Sanitization: There is no evidence of sanitization or content filtering performed on the data retrieved from the target website before it is presented to the agent.
Audit Metadata