refactor(nexus-ai): compress system prompt -600 tokens, centralize debug logging (#435, #437)

[ShunsukeHayashi]

Summary

Two cleanup tasks for the NexusAI agent:

1. System prompt compression (perf(nexus-ai): reduce system prompt token count to prevent instruction dilution #435) — reduce token count to prevent instruction dilution
2. Debug log centralization (refactor(nexus-ai): replace scattered console.log with structured debug logger #437) — replace scattered console.log with a stripped-in-production helper

Changes

System Prompt (#435)

Section	Before	After	Saved
Mermaid rules	27 lines (BAD/GOOD examples)	2 lines (essentials only)	~180 tokens
Graph schema	Full schema + semantics + examples	Reference to cypher tool	~420 tokens
Total	~2,500 tokens	~1,900 tokens	~600 tokens

The full graph schema and Cypher examples are already in the cypher tool description. Duplicating them in the system prompt wastes tokens and increases the chance of instruction dilution (the agent may follow schema examples instead of core grounding/validation rules).

Debug Logging (#437)

Before	After
4 scattered if (import.meta.env.DEV) { console.log(...) } blocks	__DEV_LOG() helper (4 calls)
Emoji in log messages (encoding risk)	Plain text prefixed with [nexus-ai]
12 lines of conditional blocks	4 single-line calls

The __DEV_LOG helper:

const __DEV_LOG = (...args: unknown[]) => {
  if (import.meta.env.DEV) console.log('[nexus-ai]', ...args);
};

Net Result

-42 lines, +14 lines = 28 fewer lines in agent.ts

Type Check

npx tsc --noEmit -p tsconfig.app.json  # No errors

Addresses #435, #437

[vercel]

@ShunsukeHayashi is attempting to deploy a commit to the NexusCore Team on Vercel.

A member of the Team first needs to authorize it.

[github-actions]

CI Report

✅ All checks passed   2365393

Pipeline

Stage	Status	Ubuntu	Windows	macOS
Typecheck	✅ success	—	—	—
Tests	✅ success	✅	✅	✅

Tests

Metric	Value
Total	3671
Passed	3651
Skipped	20
Files	1052
Duration	1m 51s

✅ All 3651 tests passed across 1052 files

20 test(s) skipped

• buildTypeEnv > known limitations (documented skip tests) > Ruby block parameter: users.each { |user| } — closure param inference, different feature
• Python match/case as-pattern type binding > resolves u.save() to User#save via match/case as-pattern binding
• Python match/case as-pattern type binding > does NOT resolve u.save() to Repo#save (negative disambiguation)
• Swift constructor-inferred type resolution > detects User and Repo classes, both with save methods
• Swift constructor-inferred type resolution > resolves user.save() to Models/User.swift via constructor-inferred type
• Swift constructor-inferred type resolution > resolves repo.save() to Models/Repo.swift via constructor-inferred type
• Swift constructor-inferred type resolution > emits exactly 2 save() CALLS edges (one per receiver type)
• Swift self resolution > detects User and Repo classes, each with a save function
• Swift self resolution > resolves self.save() inside User.process to User.save, not Repo.save
• Swift parent resolution > detects BaseModel and User classes plus Serializable protocol
• Swift parent resolution > emits EXTENDS edge: User → BaseModel
• Swift parent resolution > emits IMPLEMENTS edge: User → Serializable (protocol conformance)
• Swift cross-file User.init() inference > resolves user.save() via User.init(name:) inference
• Swift cross-file User.init() inference > resolves user.greet() via User.init(name:) inference
• Swift return type inference > detects User class and getUser function
• Swift return type inference > detects save function on User (Swift class methods are Function nodes)
• Swift return type inference > resolves user.save() to User#save via return type of getUser() -> User
• Swift return-type inference via function return type > resolves user.save() to User#save via return type of getUser()
• Swift return-type inference via function return type > user.save() does NOT resolve to Repo#save
• Swift return-type inference via function return type > resolves repo.save() to Repo#save via return type of getRepo()

Coverage

Metric	Coverage	Covered	Base (main)	Delta
Statements	68.98%	9118/13217	68.79%	📈 +0.2%
Branches	60.07%	6179/10285	59.94%	📈 +0.1%
Functions	71.63%	808/1128	71.27%	📈 +0.4%
Lines	71.22%	8127/11410	71.05%	📈 +0.2%

---

_{📋 Full run · Coverage from Ubuntu · Generated by CI}

[ShunsukeHayashi]

Closing per maintainer feedback in #467 — will resubmit one at a time after existing fix PRs are reviewed.