kb-record-decision
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands such as
grepto analyze file structure and verify plan completion. These commands use path variables derived from project files. - [PROMPT_INJECTION]: The skill demonstrates a surface for indirect prompt injection as it processes content from external plan files without sanitization.
- Ingestion points:
SKILL.mdreads project plan and checklist files from thedocs/plans/directory. - Boundary markers: None. The instructions do not define delimiters or provide warnings to ignore instructions within the processed data.
- Capability inventory: The skill performs file writes to the
.kb/decisions/directory and executes shell commands viagrep. - Sanitization: None. Content from plan files is distilled directly into new records and used in shell command strings without validation.
Audit Metadata