editor-plan-and-check
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data such as media transcripts which could potentially contain malicious instructions intended to manipulate the editor's plan.
- Ingestion points: User requests and media transcripts fetched via
timeline_describe(SKILL.md). - Boundary markers: The instructions do not mandate the use of delimiters or 'ignore' warnings for the transcript data.
- Capability inventory: The skill possesses capabilities to modify the timeline state using
timeline_applyPlanand related tools. - Sanitization: No sanitization or validation of the transcript content is specified.
- [PROMPT_INJECTION]: The skill includes instructions to suppress certain types of user-facing information, specifically regarding undo functionality.
- Evidence: 'Do not tell the user how to undo; mention ⌘Z only when they ask how to go back or are unhappy with the result.' (SKILL.md).
- Context: This is assessed as a stylistic preference for the editor's chat interface to avoid repetitive instructions, rather than a malicious concealment of activity.
Audit Metadata