mixer-policy-safety

SkillDev tools

Review, harden, and test the safety policy for AI-Controlled Party mixer/Soundcraft Ui24R scene arming. Use for approval gates, blocked mixer operations, hazardous request handling, venue validation, bypass attempts, QA, risk review, or follow-up safety changes.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the mixer-policy-safety skill

What this skill tells your AI

The instructions your AI receives, as published by pantani/tdmcp in .claude/skills/mixer-policy-safety/SKILL.md and read by ahel’s review.

Review the design from the viewpoint of a live-show safety gate.

Context to read

  • _workspace/ai-party-mixer/01_contract.md
  • _workspace/ai-party-mixer/02_adapter.md
  • current show-director schema/runtime
  • AI-Controlled Party docs/spec

Review rules

  • AI can arm; human operator approves.
  • Scene/cue execution must be predeclared and auditable.
  • mixer_gain, pa_mute, audio_routing, channel edits, and routing changes stay blocked/operator-only.
  • Live execution requires dry-run tests, adapter health, rollback/fallback notes, and a venue rehearsal.
  • Low-confidence voice/STT input cannot execute or approve anything.

Output

Write _workspace/ai-party-mixer/03_policy_qa.md with pass/fail findings, bypass attempts, required tests, validation gates, and residual risks.

Quality bar

Prefer a clear block over a clever risky shortcut. If a claim depends on venue hardware, mark it venue-validated-only.

Signals

GitHub stars
41
Forks
9
Last commit
Aug 2026
Advanced
Catalog kind
skill
Gateway key
mixer-policy-safety
Source
github.com/pantani/tdmcp
mixer-policy-safety: Skill · ahel