note
Pass
Audited by Gen Agent Trust Hub on Aug 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill facilitates an indirect prompt injection surface by automatically loading stored notes from
.rcs/notepad.mdinto the agent's context. If malicious instructions are saved to the notepad (e.g., from untrusted data sources), they could influence or compromise the agent's behavior in subsequent sessions. - Ingestion points: Content is ingested via the
/notecommand and stored in.rcs/notepad.md. - Boundary markers: The skill instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when the content is reloaded.
- Capability inventory: The skill performs local file read and write operations on
.rcs/notepad.md. - Sanitization: There is no evidence of sanitization or validation of the stored content before it is interpolated back into the agent's prompt.
Audit Metadata