expansion-signal-spotter
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from untrusted external sources which could be used to influence agent behavior.
- Ingestion points: The skill fetches data from
web_search,fetch_webpage, and external LinkedIn/Job scrapers during Phase 1 (Signal Detection). - Boundary markers: Absent. The instructions for Phase 3 (Talk Track Generation) do not include delimiters or specific instructions for the agent to ignore potential instructions embedded in the external search results or LinkedIn posts.
- Capability inventory: The skill has filesystem write access (Phase 4 saves reports to
clients/<client-name>/...) and network access via tools. - Sanitization: Absent. There is no evidence of filtering or sanitizing the content retrieved from the web before it is interpolated into prompts for generating talk tracks and expansion reports.
- [NO_CODE]: The skill consists primarily of instructions and metadata without providing any executable script or binary files, relying instead on the agent's internal reasoning and pre-installed tools.
Audit Metadata