signals-scout-general
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted event data (such as hostnames, app versions, and custom properties) from the PostHog project to generate reports and maintain a persistent scratchpad.
- Ingestion points: Data is retrieved from the project database using tools like
execute-sql,read-data-schema, and various product-specific query tools. - Boundary markers: The instructions do not define explicit delimiters or instructions to ignore potential commands embedded within the ingested event data.
- Capability inventory: The agent has the ability to author and edit reports in the project inbox (
emit-report,edit-report) and save durable observations in a scratchpad (scout-scratchpad-remember). - Sanitization: While the skill includes advanced guidance in
references/new-traffic.mdto prevent SQL injection by using hashes instead of raw strings in queries, it does not explicitly instruct the agent to sanitize or escape data before including it in the natural language reports sent to the inbox.
Audit Metadata