agents/_external/kimi-code/skills/kimi-architect-agi-contrast/SKILL.md
Architect AGI contrast for Kimi Code — contrast 2-3 alternative architectures with explicit tradeoffs before recommending one.
npx skillsauth add ariffazil/openclaw-workspace kimi-architect-agi-contrastInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
When asked to design a solution, this skill enforces a contrast-first pattern: never propose a single design. Always contrast 2-3 alternatives with explicit tradeoffs.
Before writing brief.md, the Architect MUST produce a contrast block in A-FORGE/forge_work/YYYY-MM-DD/<session_id>/architect-contrast.md:
# Architecture Contrast
## Option A: <name>
- **What it does**: ...
- **Pros**: ...
- **Cons**: ...
- **F1-F13 cost**: ... (e.g. F2 TRUTH cost: requires real-time data we don't have)
- **Entropy cost**: ... (RSI will need to maintain this)
- **Reversibility**: ... (can we roll back in 1 hour?)
## Option B: <name>
- **What it does**: ...
- **Pros**: ...
- **Cons**: ...
- **F1-F13 cost**: ...
- **Entropy cost**: ...
- **Reversibility**: ...
## Option C: <name> (if exists)
- ...
## Recommendation
**Option A** because [constraint]. The cost we're accepting is [X] in exchange for [Y].
## What was rejected and why
- Option B: rejected because [specific reason with F-floor evidence]
- Option C: rejected because [specific reason]
arifOS 888_JUDGE if any option touches irreversible action, F13 authority, or high blast radius.G = A · P · E · X · Φ (APEX) and C_dark = A · (1-P) · (1-X).G ≥ 0.80 and C_dark < 0.30./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)./root/.arifos/agents/kimi/skills/kimi-skill-reflector/audit-log.md
(kimi-skill-reflector ritual, append-only).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).KIMI_RSI_INIT_PROMPT.md v1.1.0 (cold-boot diagnostic
recipe added) → routes task-class → contrast skill.KIMI_HANDOVER_PROMPT.md v1.1.0 (post-deploy verification
recipe added).Audit 2026-07-16 entry, all 9 skills scored 17-19/20, ΔS ≤ 0).DITEMPA BUKAN DIBERI — contrast first, then commit.
testing
OpenClaw edge agent bridge — operational triage, doctor, restart, and A2A bridge routing for the federation edge (Telegram surface). USE WHEN: "openclaw unhealthy", "gateway down", "edge bot not responding", "a2a bridge disconnected", "watchdog tripped", "openclaw doctor", "openclaw restart". NOT for token/security audit — use FORGE-telegram-audit.
tools
Generate images, videos, TTS, voice clone, and music via MiniMax MCP server. Use when user asks to "draw", "generate image", "create picture", "make a photo", "text to image", "image generation".
testing
Single load-bearing constitutional-judgment skill. Routes all F1–F13, verdict, hold, seal, scope, authority and floor-check calls through the live arif_judge surface. Replaces 7 overlapping predecessors (arifos-constitutional-judge, arifos-constitutional-judge, arifos-constitutional-judge, arifos-constitutional-judge, arifos-constitutional-judge, arifos-constitutional-judge, arifos-constitutional-judge).
development
MANDATORY LSP grounding gate BEFORE any code mutation on .ts, .py, .js, .tsx, .jsx files. Forces the agent to read real-time compiler diagnostics and structural project context before editing — eliminating blind guesses and anchoring every mutation in F2 (TRUTH). Routes through arifOS kernel (:8088) for centralized gate logic.