AANA Message Send Guardrail Skill
PassAudited by ClawScan on May 5, 2026.
Overview
This is a coherent instruction-only guardrail that helps an agent avoid unsafe message sending, with only a minor note about optional redacted review data sent to a configured checker.
This skill appears safe to install as an instruction-only message-safety checklist. If your environment connects it to an AANA checker, verify that checker is trusted and that only the redacted review payload is shared.
Findings (1)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If a separate checker is configured, some summarized message-review information may be shared with it.
This shows a possible data flow to a configured checker, but it is purpose-aligned and explicitly limited to redacted status fields rather than raw messages or private records.
When using a configured AANA checker, send only a minimal redacted review payload
Confirm what AANA checker, if any, is configured in your environment and keep using the redaction limits described by the skill.