journal-selector
Pass
Audited by Gen Agent Trust Hub on Jul 26, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill utilizes user-provided journal names to resolve and load local configuration files (e.g.,
journals/<folder>/SKILL.md). This introduces a surface for indirect prompt injection where instructions in the resolved files could influence the agent's logic and formatting tasks.\n - Ingestion points: User input for journal names or abbreviations is accepted in the
Behaviorsection ofSKILL.md.\n - Boundary markers: Absent. The skill does not instruct the agent to use delimiters or ignore instructions found within the loaded journal files.\n
- Capability inventory: The skill reads local files from the
journals/directory and passes extracted parameters to downstream tools likefood-pipelineandfood-figure.\n - Sanitization: No explicit sanitization or validation of the user input or the resolved file content is defined.
Audit Metadata