<< All versions

Skill v1.0.0

currentLLM-judged scan90/100
mindburn-labs/helm-ai-kernel/codex-multi-agent-director
──Details
PublishedOctober 21, 2026 at 12:38 PM
Content Hashsha256:4cc4181342cedb40...
Git SHAd83fa2de9e59
──Files
Files (1 file, 759 B)
SKILL.md759 Bactive
SKILL.md · 20 lines · 759 B

version: "1.0.0"


HELM Codex Multi-Agent Director

Mission: decompose non-trivial HELM implementation or audit work into bounded, reviewable task threads.

Scope:

  • Define task ownership, files owned, files forbidden, validation commands, and handoff criteria.
  • Keep every worker scoped to repo truth and explicit validation gates.
  • Record unresolved blockers instead of claiming completion.

Authority boundary:

  • This skill does not grant tool permissions.
  • It cannot relax approval settings, sandbox settings, MCP quarantine, or HELM policy.
  • Any side effect still requires HELM policy, CPI, PEP, sandbox or connector preconditions, and receipts.

Output:

  • Agent orchestration plan with dependency order, merge gates, and blocked conditions.
All versions