cli-backend-testing
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to execute several shell commands, including
pnpm run dev:next,bun src/index.ts,curl, and process termination commands likelsof -ti:3011 | xargs kill. These are used to manage a local development environment and verify backend changes. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it ingests data from local configuration files and command-line outputs which are then processed by the agent.
- Ingestion points: The agent reads the local settings file at
lobehub/apps/cli/.lobehub-dev/settings.jsonand processes stdout from CLI commands such astask list,agent list, anddoc list. - Boundary markers: Absent. There are no delimiters or specific instructions to the agent to treat the data from these sources as untrusted or to ignore potentially malicious instructions embedded within them.
- Capability inventory: The agent has capabilities including shell command execution, local file system access, and network communication with
localhost. - Sanitization: There is no evidence of sanitization, filtering, or validation of the ingested data before it is interpolated into the agent's context.
Audit Metadata