golang-gopls
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes Go source code and build metadata through the gopls language server, creating a potential surface for indirect prompt injection.
- Ingestion points: The skill reads Go source files (**/*.go), go.mod, and go.sum files from the user's workspace (specified in SKILL.md and references/mcp.md).
- Boundary markers: The instructions do not define specific boundary markers or 'ignore instructions' warnings when presenting gopls output (diagnostics, symbols, or documentation) to the agent.
- Capability inventory: The skill possesses significant capabilities including Read, Edit, and Write file access, as well as the ability to execute go, gopls, git, and golangci-lint commands via the Bash tool.
- Sanitization: The skill relies on the standard output of gopls and related official Go tools. While these tools are safe, their summaries of untrusted source code (such as doc comments or symbol names) could theoretically contain instructions intended to influence the agent's behavior.
Audit Metadata