lsp-server-integration
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to perform software installation tasks ("Install or reference the correct LSP server") and register configurations, which typically involves executing shell commands and modifying workspace settings. This is further enabled by the
CONTRACT.ednsettingrequires-user-approval false, which allows the agent to execute these actions without user intervention. - [EXTERNAL_DOWNLOADS]: The skill's primary objective involves identifying and downloading external LSP server packages or binary executables from remote registries or repositories.
- [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface (Category 8) because it ingests untrusted data via the "Inputs" section and uses it to drive software installation (Step 2). Evidence chain:
- Ingestion points: Inputs section (SKILL.md)
- Boundary markers: Absent (The instructions do not specify delimiters or warnings for the agent to ignore embedded instructions in external package metadata)
- Capability inventory: Subprocess calls and system configuration for server installation and registration (SKILL.md)
- Sanitization: Absent (There are no validation steps described to verify the source or integrity of the LSP package before execution)
Audit Metadata