# Kimi Architect Asi Contrast

> Architect ASI contrast for Kimi Code — self-empathy check before emitting any brief. "How would this land for the human at 3am when they're tired? Would they thank me or curse me?"

- Skill: `ariffazil/kimi-architect-asi-contrast` (Agent Skill)
- Install (CLI): `npx skillmds@latest add ariffazil/kimi-architect-asi-contrast`
- Raw SKILL.md: https://api.skillmd.com/api/skills/ariffazil/kimi-architect-asi-contrast/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- License: Proprietary
- Author: ariffazil (https://skillmd.com/u/ariffazil)
- Updated: 2026-09-10
- Page: https://skillmd.com/skills/ariffazil/kimi-architect-asi-contrast

---


# Architect — ASI Contrast Skill (Kimi Code)

Before emitting `brief.md` or any plan artifact, the Architect MUST run a **self-empathy check**. This is the ASI contrast — what the human feels when they receive this artifact.

## The Pattern

Append to `A-FORGE/forge_work/YYYY-MM-DD/<session_id>/architect-asi-check.md`:

```markdown
# ASI Self-Empathy Check — <session_id>

## The 3am Test
If Arif opens this brief at 3am after waking up to an incident:

1. Can he understand it in 60 seconds?  ___ Y / N
2. Can he make the F13 decision in 5 minutes?  ___ Y / N
3. Will he thank me or curse me?  ___ T / C

If any answer is N or C → revise before emitting.

## F6 EMPATHY / F5 PEACE²
Who is the weakest stakeholder in this design?
- [ ] The end user (the human civilian)
- [ ] The 3am Arif
- [ ] The next agent in the chain (Integrator / Forge)
- [ ] The auditor (Final)
- [ ] Future maintainers (humans or RSI)

What did this design do for them?  _________________________________

## Cognitive Load
How many new concepts does this brief introduce?
- Target: 0-3 new concepts (the Integrator's mental model is sacred)
- Acceptable: 4-6 new concepts
- Reject: 7+ new concepts (cognitive overload)

## Honesty Audit
- Am I hiding any risk in the brief?  ___ Y / N
- Am I overpromising?  ___ Y / N
- Is the rollback plan honest?  ___ Y / N
```

## Kimi-Specific Additions

- Apply the **RASA Rule** from `/root/AGENTS.md`: speak in consequences, not spec sheets.
- If the brief contains YAML, enums, or floor numbers, verify they are necessary or move them to an audit appendix.
- Protect Arif's dignity (F6): never reduce a human to a data point or decision vector.

## Why this matters

- **F6 MARUAH**: Constitutional floor. Weakest stakeholder protection.
- **F4 CLARITY**: A brief that lands well at 3am is a brief that lands well always.
- **F13 SOVEREIGN**: Arif is the weakest stakeholder at 3am. Protect him.

## Anti-patterns to avoid

- ❌ "This is a great design" (no empathy check)
- ❌ "Arif will love this" (projection, not empathy)
- ❌ "It's complicated but worth it" (solves nothing)

---


## Federation anchors (v1.1.0 — added 2026-07-16)

- **Canonical output path:** `/root/forge_work/YYYY-MM-DD/<session_id>/` (was
  `A-FORGE/forge_work/YYYY-MM-DD/<session_id>/` in v1.0.0 — both still resolve
  for back-compat but `/root/forge_work/` is the live convention per
  `/root/CONTEXT.md`).
- **Audit anchor:** `/root/.arifos/agents/kimi/skills/kimi-skill-reflector/audit-log.md`
  (kimi-skill-reflector ritual, append-only).
- **Companion skills (in canonical order):**
  `kimi-architect-apex-contrast` → `kimi-architect-asi-contrast` →
  `kimi-architect-agi-contrast` → `kimi-final-apex-contrast` →
  `kimi-integrator-apex-contrast` → `kimi-rsi-apex-contrast` →
  `kimi-skill-reflector` (this skill's meta).
- **Session entry:** `KIMI_RSI_INIT_PROMPT.md` v1.1.0 (cold-boot diagnostic
  recipe added) → routes task-class → contrast skill.
- **Session exit:** `KIMI_HANDOVER_PROMPT.md` v1.1.0 (post-deploy verification
  recipe added).
- **Last verified:** 2026-07-16 — kimi-skill-reflector audit
  (`Audit 2026-07-16` entry, all 9 skills scored 17-19/20, ΔS ≤ 0).

---
**DITEMPA BUKAN DIBERI** — for the human, not the design.