game-build-log-triage
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to its core function of processing untrusted log data while possessing active capabilities.
- Ingestion points: Untrusted data enters the agent context through the analysis of raw log text, Editor.log files, BuildCookRun logs, and CI transcripts provided by the user (documented in SKILL.md Step 1).
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for the processed log content, increasing the risk of the agent following instructions embedded in malicious logs.
- Capability inventory: The skill is granted access to the
Bash,Write, andEdittools in addition to read-only tools (documented in theallowed-toolsfrontmatter of SKILL.md). - Sanitization: There is no requirement for validation or sanitization of external log content before it is processed for triage.
Audit Metadata