go-error-modernization
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests untrusted source code, exposing it to indirect prompt injection. 1. Ingestion points: Go files (.go) and modules (go.mod) processed via find and grep. 2. Boundary markers: Detailed decision tree and mandated suppression reasons in code comments. 3. Capability inventory: Access to bash, go, edit, and grep tools for analysis and modification. 4. Sanitization: Relies on the agent following instructions and manual verification through compilation and testing.
- [COMMAND_EXECUTION]: The skill uses shell commands to execute 'erraudit', a tool provided by the author, to automate code transformations. Evidence: 'GOEXPERIMENT=jsonv2 erraudit fix ./...'.
- [SAFE]: References to standard Go documentation and the author's public repositories are used for technical context. Evidence: Links to go.dev and github.com/larsartmann/go-error-family.
Audit Metadata