Security audit
Security checks for vulnerabilities and agentic risk
The skill provides coherent Nansen research guidance with disclosed account access, usage costs, and feedback controlled by your saved preference or explicit approval.
Research commands use your Nansen account and may consume credits. Review your saved feedback preference: enabling it permits concise product reports without approval each time. Leave it disabled if you want to review every report.
The supplied shell commands perform specific data queries. They contain no routing guidance, decision logic, or use of nansen agent. Although the commands belong to the Nansen research domain, their primary behavior materially differs from the declared purpose of helping choose a research approach.
The manifest describes deciding between AI research and direct data commands, but this section instructs the agent to inspect saved feedback preferences and submit product-gap reports, potentially without asking again when a preference is enabled. This is a separate account-integrated reporting workflow, not an implementation detail of research routing; its consent safeguards do not resolve the manifest-level scope mismatch.
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
When a Nansen task exposes a useful product gap, draft a concise report without
secrets or unnecessary conversation details. Before sending on your own, read
the current account's saved feedback preference with
`nansen feedback preferences --json`. Use a fresh read for each new report. If enabled, you may send without asking again,
subject to the host's controls. If off, unset or unavailable, show the proposed
report and ask the user. If they decline or you cannot ask, do not send.
No suspicious patterns detected.