signals-scout-feature-flags

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security vulnerabilities were detected in the skill instructions or logic.
  • [PROMPT_INJECTION]: The skill demonstrates high security awareness by including a dedicated section on handling untrusted data. It explicitly warns the agent that event-supplied strings (like flag keys or responses) could be crafted as instructions and provides clear mitigation strategies: treating them strictly as data, sanitizing/truncating snippets, and ensuring event values never authorize actions.
  • [DATA_EXFILTRATION]: All data access is performed using specialized PostHog MCP tools (e.g., feature-flag-get-all, execute-sql) within the authorized project scope. The skill emits findings through a controlled signal emission tool (signals-scout-emit-signal) rather than arbitrary network requests.
  • [COMMAND_EXECUTION]: The provided SQL queries for traffic analysis are read-only and target specific analytics tables (events, system.feature_flags). They do not involve OS-level command execution or unsafe interpolation of user input into shell environments.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 12:35 PM
Security Audit — agent-trust-hub — signals-scout-feature-flags