# Loki Mode

> ALWAYS use this when the request matches Loki Mode: Version 2.35.0 | PRD to Production | Zero Human Intervention > Research-enhanced: OpenAI SDK, DeepMind, Anthropic, AWS Bedrock, Agent SDK, HN Production (2025)

- Skill: `anhvu1107/loki-mode` (Agent Skill, multi-file: 1105 files)
- Install (CLI): `npx skillmds@latest add anhvu1107/loki-mode`
- Raw SKILL.md: https://api.skillmd.com/api/skills/anhvu1107/loki-mode/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Product & Planning
- Author: anhvu1107 (https://skillmd.com/u/anhvu1107)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/anhvu1107/loki-mode

---


# Loki Mode - Multi-Agent Autonomous Startup System

## Selective Reading Rule

Start with:

- `references/senior-master-standard.md`
- `references/usage-routing.md`
- `references/quality-checklist.md`

Then load only the inherited docs, scripts, assets, examples, or detailed instructions that match the user's actual task.

## Progressive Disclosure

- Read `references/full-skill-instructions.md` only when the task needs the full original upstream guidance.
- Prefer bundled scripts for deterministic or repetitive operations when they exist.
- Reuse bundled assets instead of inventing substitutes when assets are present.
- Keep final answers aligned with this skill's frontmatter trigger, Senior Master standard, and Codex strict review gate.