feature-availability

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to process external 'signal content' (such as product updates or Slack messages) to extract feature names. This creates a surface for indirect prompt injection where malicious instructions embedded in the signals could attempt to influence agent behavior.
  • Ingestion points: Signal content processed during the 'Extract Feature Mentions' step in SKILL.md.
  • Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the signal data.
  • Capability inventory: The skill utilizes posthog MCP tools for data retrieval and performs file system write operations to generate reports and maintain a local cache.
  • Sanitization: There is no evidence of sanitization or validation of the input signal content before processing.
  • [COMMAND_EXECUTION]: The skill defines local file paths for report generation (e.g., pm-workspace-docs/status/) and caching. These are standard operations for the intended workflow and do not involve shell execution or elevated privileges.
  • [CREDENTIALS_UNSAFE]: A hardcoded PostHog Project ID (81505) is present in the configuration. This is a non-sensitive identifier used to scope API queries and does not constitute a secret or credential exposure.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 05:31 PM