input-work-evidence-explorer

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements strong privacy boundaries, explicitly instructing the agent to never inspect home directories, cloud accounts, or broad repository collections without specific, narrow authorization.
  • [SAFE]: It utilizes a 'governed-import' or 'session-only' mode to manage data persistence and ensures that sensitive materials (secrets, binaries, dependencies) are skipped during inspection.
  • [SAFE]: References to external skills (e.g., evidence-led-interviewer, output-career-portfolio) follow a modular, internal design and do not involve untrusted remote code execution.
  • [SAFE]: The why-hire-me-update check is described as a cached, read-only version check that requires user agreement before updating, which is a standard and safe software maintenance pattern.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 09:16 PM
Security Audit — agent-trust-hub — input-work-evidence-explorer