shipyard-debugging
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides examples of shell commands, including
security list-keychains,codesign, andenv, as part of a diagnostic instrumentation strategy for debugging multi-component systems like build and signing pipelines. - [INDIRECT_PROMPT_INJECTION]: The skill's core function involves processing untrusted external data, such as error messages, stack traces, and test logs, which could serve as a surface for indirect prompt injection attacks.
- Ingestion points: Error messages, tracebacks, and test failures (SKILL.md, Phase 1).
- Boundary markers: None defined; instructions encourage reading error messages "carefully" and "completely."
- Capability inventory: The skill demonstrates capabilities for shell command execution (Phase 1, Step 4) and file system/git access (Phase 1, Step 3).
- Sanitization: No specific sanitization or filtering of external error data is prescribed.
Audit Metadata