dev-event-kickoff

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a state-management system using a local markdown file (event-context.md) to maintain context between user sessions. This allows the agent to resume work without re-interviewing the user.
  • [SAFE]: The skill includes a 'Detect before asking' phase that inventories local project files (e.g., README, budget spreadsheets, venue contracts) and reads git history to infer the current project stage. This is a standard context-gathering behavior for developer-oriented AI agents.
  • [SAFE]: The instructions contain explicit security guardrails regarding data privacy. It mandates that personal data (PII), code-of-conduct reports, and confidential negotiated rates must never be stored in persistent memory.
  • [SAFE]: All task routing targets the same vendor's collection (samber/dev-event-organizer-skills) or recommends sibling collections for non-event tasks, ensuring a consistent and expected execution environment.
  • [SAFE]: The skill's persistence mechanism (updating local instruction files with project invariants) is a documented feature of the environment used to ensure consistent behavior across sessions without requiring repeated user input.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 09:13 PM
Security Audit — agent-trust-hub — dev-event-kickoff