tella-screen-callouts
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input that could contain malicious instructions designed to manipulate agent behavior.
- Ingestion points: Accesses video transcripts and interaction events through tools like
get_timeline,get_mouse_events, andget_key_events(mentioned in SKILL.md). - Boundary markers: The skill lacks explicit markers or instructions to isolate ingested data from the core logic.
- Capability inventory: Employs Tella MCP tools like
add_highlightto modify video projects. No system-level or network-exfiltration capabilities are present. - Sanitization: There is no evidence of data sanitization or filtering for the processed transcripts or logs.
Audit Metadata