exploring-scouts
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for read-only exploration of PostHog Signals scouts. It uses official MCP tools to query vendor-managed diagnostic data from PostHog infrastructure (us.posthog.com). No credentials or sensitive local data are accessed or transmitted to third parties.
- [COMMAND_EXECUTION]: The skill provides three Python scripts (
scripts/assess_health.py,scripts/fleet_survey.py,scripts/render_run_report.py) for local data processing. These tools are used to format large JSON payloads, such as diagnostic logs, which would otherwise exceed token limits during interaction. Analysis of these scripts confirms they use only standard Python libraries and perform no network I/O or unauthorized filesystem access. - [PROMPT_INJECTION]: The skill ingests and displays content that could be influenced by other automated agents, such as scout run summaries and scratchpad memory. While this represents a surface for indirect prompt injection, the skill's purpose is exploratory and read-only, and it does not use the ingested data to trigger automated write operations or execute code.
Audit Metadata