context-journal-potato
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it summarizes session data which could contain malicious instructions.
- Ingestion points: The agent reads the current session's context to identify through-lines and 'diamonds' for the journal entry (SKILL.md).
- Boundary markers: There are no instructions or delimiters defining how the agent should distinguish between legitimate session content and potential embedded instructions.
- Capability inventory: The skill possesses the capability to write files to the local filesystem (SKILL.md, Procedure step 5).
- Sanitization: The skill does not specify any sanitization, escaping, or validation of the summarized content before it is written to disk.
- [COMMAND_EXECUTION]: The skill involves file system operations to store the generated journal entries.
- The agent is instructed to write a Markdown file to the family's 'out/' directory or a user-provided override path (SKILL.md).
Audit Metadata