knock-in-app-ui
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests JSON data from the Knock API via CLI commands and uses it to generate React components and shell commands. Ingestion points: Data from
knock guide listandknock message-type get. Boundary markers: Absent; no delimiters or ignore instructions are provided for the API output. Capability inventory: Shell command execution (knock,jq,rm), file system modification (.env), and code generation. Sanitization: Absent; the agent is not instructed to validate or escape data from the remote service. - [COMMAND_EXECUTION]: Executes shell commands including the
knockCLI for environment and resource management, andrm -rffor local scaffolding cleanup. - [EXTERNAL_DOWNLOADS]: Installs the
@knocklabs/reactlibrary from the public NPM registry. This is a legitimate vendor package required for the integration.
Audit Metadata