config-audit/tests/scanners/token-hotspots.test.mjs

import { describe, it, beforeEach } from 'node:test';
import assert from 'node:assert/strict';
import { resolve } from 'node:path';
import { fileURLToPath } from 'node:url';
import { resetCounter } from '../../scanners/lib/output.mjs';
import { scan } from '../../scanners/token-hotspots.mjs';
import { discoverConfigFiles } from '../../scanners/lib/file-discovery.mjs';
import { withHermeticHome } from '../helpers/hermetic-home.mjs';

const __dirname = fileURLToPath(new URL('.', import.meta.url));
const FIXTURES = resolve(__dirname, '../fixtures');

async function fixtureDiscovery(name) {
  return discoverConfigFiles(resolve(FIXTURES, name));
}

async function runScanner(fixtureName) {
  resetCounter();
  const path = resolve(FIXTURES, fixtureName);
  const discovery = await fixtureDiscovery(fixtureName);
  // Hermetic HOME: scan() calls readActiveConfig, which resolves the user
  // ~/.claude/CLAUDE.md cascade + plugins/MCP from process.env.HOME. Without
  // this override the developer's real ~/.claude leaks into every fixture
  // result — small-cascade tips past the 10k-token threshold and sonnet-era
  // picks up ambient MCP findings — making assertions machine-dependent
  // (green on a clean HOME, red on a populated one). Mirrors the OST test.
  // NOTE: residual coupling — the cascade's ancestor walk still climbs from
  // the fixture path through the real repo CLAUDE.md (an ancestor of tests/),
  // but that contribution is deterministic per-checkout, not user-dependent.
  return withHermeticHome(() => scan(path, discovery));
}

describe('TOK scanner — healthy-project', () => {
  let result;
  beforeEach(async () => {
    result = await runScanner('healthy-project');
  });

  it('returns status ok', () => {
    assert.equal(result.status, 'ok');
  });

  it('reports scanner prefix TOK', () => {
    assert.equal(result.scanner, 'TOK');
  });

  it('finding IDs match CA-TOK-NNN pattern', () => {
    for (const f of result.findings) {
      assert.match(f.id, /^CA-TOK-\d{3}$/);
    }
  });

  it('exposes total_estimated_tokens as a number', () => {
    assert.equal(typeof result.total_estimated_tokens, 'number');
    assert.ok(result.total_estimated_tokens >= 0);
  });
});

describe('TOK scanner — opus-47/cache-breaking', () => {
  let result;
  beforeEach(async () => {
    result = await runScanner('opus-47/cache-breaking');
  });

  it('flags CA-TOK-001 (cache-breaking volatile top)', () => {
    const f = result.findings.find(x => x.id === 'CA-TOK-001');
    assert.ok(f, 'expected a CA-TOK-001 finding for cache-breaking fixture');
  });

  it('CA-TOK-001 severity is high (v5 F7 recalibration)', () => {
    const f = result.findings.find(x => x.id === 'CA-TOK-001');
    assert.equal(f.severity, 'high', `expected high after F7, got ${f.severity}`);
  });

  it('CA-TOK-001 description carries no stale Opus-4.7 anchor (mechanic is model-agnostic)', () => {
    const f = result.findings.find(x => x.id === 'CA-TOK-001');
    assert.doesNotMatch(f.description, /Opus 4\.7/,
      'prompt-cache mechanics are a feature property, not model-specific; do not era-stamp a stale model');
  });
});

describe('TOK scanner — opus-47/redundant-tools', () => {
  let result;
  beforeEach(async () => {
    result = await runScanner('opus-47/redundant-tools');
  });

  it('emits at least one CA-TOK-002 finding (redundant tool/permission)', () => {
    const has002 = result.findings.some(f => /^CA-TOK-002$/.test(f.id) || f.title?.toLowerCase().includes('redundant'));
    assert.ok(has002, 'expected a CA-TOK-002 finding for redundant-tools fixture');
  });
});

describe('TOK scanner — opus-47/deep-imports', () => {
  let result;
  beforeEach(async () => {
    result = await runScanner('opus-47/deep-imports');
  });

  it('emits at least one CA-TOK-003 finding (deep @import chain)', () => {
    const has003 = result.findings.some(f => /^CA-TOK-003$/.test(f.id) || f.title?.toLowerCase().includes('import'));
    assert.ok(has003, 'expected a CA-TOK-003 finding for deep-imports fixture');
  });
});

describe('TOK scanner — opus-47/sonnet-era (v5 F5: Pattern D removed)', () => {
  let result;
  beforeEach(async () => {
    result = await runScanner('opus-47/sonnet-era');
  });

  it('emits zero findings (no Pattern D / CA-TOK-004 anymore)', () => {
    assert.equal(result.findings.length, 0,
      `expected 0 findings on sonnet-era after F5, got: ${result.findings.map(f => f.id).join(', ')}`);
  });

  it('never emits CA-TOK-004 (removed in v5)', () => {
    assert.ok(result.findings.every(f => f.id !== 'CA-TOK-004'),
      'expected no CA-TOK-004; removed in v5 F5');
  });
});

describe('TOK scanner — marketplace scale ordering', () => {
  it('total_estimated_tokens strictly increases across small → medium → large', async () => {
    const small = await runScanner('marketplace-small');
    const medium = await runScanner('marketplace-medium');
    const large = await runScanner('marketplace-large');

    assert.ok(small.total_estimated_tokens < medium.total_estimated_tokens,
      `expected small (${small.total_estimated_tokens}) < medium (${medium.total_estimated_tokens})`);
    assert.ok(medium.total_estimated_tokens < large.total_estimated_tokens,
      `expected medium (${medium.total_estimated_tokens}) < large (${large.total_estimated_tokens})`);
  });
});

describe('TOK scanner — readActiveConfig integration (v5 F1)', () => {
  let result;
  beforeEach(async () => {
    result = await runScanner('tok-active-config');
  });

  it('exposes activeConfig summary on the result (proves readActiveConfig was called)', () => {
    assert.ok(result.activeConfig, 'expected result.activeConfig to be set');
    assert.equal(typeof result.activeConfig.claudeMdEstimatedTokens, 'number');
    assert.ok(result.activeConfig.claudeMdEstimatedTokens > 0,
      `expected claudeMd cascade > 0 tokens, got ${result.activeConfig.claudeMdEstimatedTokens}`);
  });

  it('hotspots include at least one MCP-source entry', () => {
    const hasMcp = result.hotspots.some(h => /mcp/i.test(h.source));
    assert.ok(hasMcp,
      `expected hotspots to include an MCP source; got: ${result.hotspots.map(h => h.source).join(', ')}`);
  });

  it('total_estimated_tokens exceeds the minimal sonnet-era baseline', async () => {
    // sonnet-era has no .mcp.json — the activeConfig MCP entries from this
    // fixture should push its total above sonnet-era's even when both fixtures
    // share the user's ambient cascade/plugin state.
    const baseline = await runScanner('opus-47/sonnet-era');
    assert.ok(result.total_estimated_tokens > baseline.total_estimated_tokens,
      `expected ${result.total_estimated_tokens} > ${baseline.total_estimated_tokens}`);
  });
});

describe('TOK scanner — hotspots contract', () => {
  let result;
  beforeEach(async () => {
    result = await runScanner('marketplace-large');
  });

  it('every finding has a non-empty recommendation', () => {
    for (const f of result.findings) {
      assert.ok(f.recommendation, `finding ${f.id} missing recommendation`);
      assert.ok(String(f.recommendation).length > 0, `finding ${f.id} has empty recommendation`);
    }
  });

  it('exposes a hotspots array of length 3–10', () => {
    assert.ok(Array.isArray(result.hotspots), 'expected result.hotspots to be an array');
    assert.ok(result.hotspots.length >= 3, `expected ≥3 hotspots, got ${result.hotspots.length}`);
    assert.ok(result.hotspots.length <= 10, `expected ≤10 hotspots, got ${result.hotspots.length}`);
  });

  it('every hotspot exposes source/estimated_tokens/rank/recommendations', () => {
    for (const h of result.hotspots) {
      assert.ok(typeof h.source === 'string' && h.source.length > 0, 'hotspot.source missing');
      assert.equal(typeof h.estimated_tokens, 'number', 'hotspot.estimated_tokens not a number');
      assert.equal(typeof h.rank, 'number', 'hotspot.rank not a number');
      assert.ok(Array.isArray(h.recommendations), 'hotspot.recommendations not an array');
      assert.ok(h.recommendations.length >= 1 && h.recommendations.length <= 3,
        `hotspot.recommendations length should be 1–3, got ${h.recommendations.length}`);
    }
  });

  it('every hotspot carries a load-pattern triple (v5.6 B2)', () => {
    const LOAD_PATTERNS = new Set(['always', 'on-demand', 'external', 'unknown']);
    const SURVIVES = new Set(['yes', 'no', 'n/a']);
    for (const h of result.hotspots) {
      assert.ok(LOAD_PATTERNS.has(h.loadPattern),
        `hotspot ${h.source} has invalid loadPattern: ${h.loadPattern}`);
      assert.ok(SURVIVES.has(h.survivesCompaction),
        `hotspot ${h.source} has invalid survivesCompaction: ${h.survivesCompaction}`);
      assert.ok(typeof h.derivationConfidence === 'string' && h.derivationConfidence.length > 0,
        `hotspot ${h.source} missing derivationConfidence`);
    }
  });

  it('tags a CLAUDE.md hotspot always-loaded and a skill hotspot on-demand (v5.6 B2)', () => {
    const claudeMd = result.hotspots.find(h => /CLAUDE\.md/.test(h.source));
    if (claudeMd) {
      assert.equal(claudeMd.loadPattern, 'always',
        `expected CLAUDE.md hotspot always-loaded; got ${claudeMd.loadPattern}`);
    }
    const skill = result.hotspots.find(h => /SKILL\.md/.test(h.source) || /skills?\//.test(h.source));
    if (skill) {
      assert.equal(skill.loadPattern, 'on-demand',
        `expected skill hotspot on-demand; got ${skill.loadPattern}`);
    }
  });

  it('every hotspot.source is unique (v5 F4: no padding)', () => {
    const sources = result.hotspots.map(h => h.source);
    const unique = new Set(sources);
    assert.equal(unique.size, sources.length,
      `expected unique sources; got duplicates in: ${sources.join(', ')}`);
  });

  it('hotspots.length never exceeds HOTSPOTS_MAX (10)', () => {
    assert.ok(result.hotspots.length <= 10,
      `expected ≤10 hotspots, got ${result.hotspots.length}`);
  });
});

describe('TOK scanner — M2 skill description > 500 chars (v5)', () => {
  it('flags skill with bloated description (low severity)', async () => {
    const result = await runScanner('skill-bloated');
    const f = result.findings.find(x => /skill description/i.test(x.title || ''));
    assert.ok(f, `expected skill-description finding; got: ${result.findings.map(x => x.title).join(' | ')}`);
    assert.equal(f.severity, 'low', `expected low, got ${f.severity}`);
    assert.match(f.evidence || '', /bloated/);
  });

  it('does NOT flag tight description (under 500 chars)', async () => {
    const result = await runScanner('skill-tight');
    const f = result.findings.find(x => /skill description/i.test(x.title || ''));
    assert.equal(f, undefined, `expected no skill-description finding; got: ${f?.title}`);
  });
});

describe('TOK scanner — M4 cascade > 10k tokens (v5)', () => {
  it('flags CLAUDE.md cascade > 10k tokens with medium severity', async () => {
    const result = await runScanner('large-cascade');
    const f = result.findings.find(x => /cascade/i.test(x.title || ''));
    assert.ok(f, `expected cascade finding; got: ${result.findings.map(x => x.title).join(' | ')}`);
    assert.equal(f.severity, 'medium', `expected medium, got ${f.severity}`);
    assert.match(f.title, /CLAUDE\.md cascade/i);
  });

  it('does NOT flag small cascade (< 10k tokens)', async () => {
    const result = await runScanner('small-cascade');
    const f = result.findings.find(x => /cascade/i.test(x.title || ''));
    assert.equal(f, undefined,
      `expected no cascade finding for small fixture; got: ${f?.title}`);
  });
});

describe('TOK scanner — N1 MCP tool-schema budget (v5 CA-TOK-005)', () => {
  // readActiveConfig pulls in ambient ~/.claude.json plugin MCP servers; tests
  // filter to the fixture's own server name (budget-srv-<count>) to avoid
  // user-state leakage. Findings identified by title (not exact ID) — TOK IDs
  // are sequential per scan.
  const findFixtureBudget = (result, count) =>
    result.findings.find(f =>
      /MCP tool-schema budget/i.test(f.title || '') &&
      (f.title || '').includes(`budget-srv-${count}`)
    );

  it('14 tools → no budget finding (under 20-tool floor)', async () => {
    const result = await runScanner('mcp-budget/14-tools');
    const f = findFixtureBudget(result, 14);
    assert.equal(f, undefined,
      `expected no budget finding for budget-srv-14 under 20 tools; got: ${f?.title}`);
  });

  it('25 tools → low severity', async () => {
    const result = await runScanner('mcp-budget/25-tools');
    const f = findFixtureBudget(result, 25);
    assert.ok(f, `expected budget finding for budget-srv-25; got: ${result.findings.map(x => x.title).join(' | ')}`);
    assert.equal(f.severity, 'low', `expected low for 25 tools, got ${f.severity}`);
  });

  it('60 tools → medium severity', async () => {
    const result = await runScanner('mcp-budget/60-tools');
    const f = findFixtureBudget(result, 60);
    assert.ok(f, `expected budget finding for budget-srv-60`);
    assert.equal(f.severity, 'medium', `expected medium for 60 tools, got ${f.severity}`);
  });

  it('120 tools → high severity', async () => {
    const result = await runScanner('mcp-budget/120-tools');
    const f = findFixtureBudget(result, 120);
    assert.ok(f, `expected budget finding for budget-srv-120`);
    assert.equal(f.severity, 'high', `expected high for 120 tools, got ${f.severity}`);
  });

  it('unknown toolCount → low severity with "unknown" in evidence', async () => {
    const result = await runScanner('mcp-budget/unknown-tools');
    const f = findFixtureBudget(result, 'unknown');
    assert.ok(f, `expected budget finding for budget-srv-unknown`);
    assert.equal(f.severity, 'low', `expected low for unknown toolCount, got ${f.severity}`);
    assert.match(String(f.evidence || ''), /unknown/i,
      `expected "unknown" in evidence, got: ${f.evidence}`);
  });

  it('finding ID matches CA-TOK-NNN format', async () => {
    const result = await runScanner('mcp-budget/120-tools');
    const f = findFixtureBudget(result, 120);
    assert.ok(f);
    assert.match(f.id, /^CA-TOK-\d{3}$/);
  });

  it('finding evidence carries calibration_note', async () => {
    const result = await runScanner('mcp-budget/60-tools');
    const f = findFixtureBudget(result, 60);
    assert.ok(f);
    assert.match(String(f.evidence || ''), /severity reflects estimated tokens\/turn/i);
  });
});

describe('TOK scanner — F7 severity recalibration (v5)', () => {
  // Findings identified by title pattern, not finding ID — TOK IDs are
  // sequential per scan run, not semantic per pattern (output.mjs:31).
  const SEVERITY_TABLE = [
    { fixture: 'opus-47/cache-breaking',  pattern: 'A', titleMatch: /cache-breaking volatile/i, expected: 'high' },
    { fixture: 'opus-47/redundant-tools', pattern: 'B', titleMatch: /redundant permission/i,    expected: 'medium' },
    { fixture: 'opus-47/deep-imports',    pattern: 'C', titleMatch: /deep @import chain/i,      expected: 'low' },
  ];

  for (const { fixture, pattern, titleMatch, expected } of SEVERITY_TABLE) {
    it(`Pattern ${pattern} (${fixture}) has severity ${expected}`, async () => {
      const result = await runScanner(fixture);
      const f = result.findings.find(x => titleMatch.test(x.title || ''));
      assert.ok(f, `expected a finding matching ${titleMatch} in ${fixture}; got: ${result.findings.map(x => x.title).join(' | ')}`);
      assert.equal(f.severity, expected, `expected ${expected}, got ${f.severity}`);
    });

    it(`Pattern ${pattern} (${fixture}) carries calibration_note evidence`, async () => {
      const result = await runScanner(fixture);
      const f = result.findings.find(x => titleMatch.test(x.title || ''));
      assert.ok(f, `expected a finding matching ${titleMatch} in ${fixture}`);
      const evidence = String(f.evidence || '');
      assert.ok(/severity reflects estimated tokens\/turn/i.test(evidence),
        `expected calibration_note phrase in evidence, got: ${evidence}`);
    });
  }
});

// ── Pattern H: stale plugin-cache versions (v5.9 B3) ──────────────────────
describe('TOK scanner — H stale plugin-cache versions (v5.9 B3)', () => {
  const staleDiscovery = {
    files: [],
    staleCacheVersions: [
      { key: 'mkt/voyage/5.1.1', fileCount: 35, estimatedBytes: 416237 },
      { key: 'mkt/config-audit/5.6.0', fileCount: 149, estimatedBytes: 301330 },
    ],
  };

  it('emits a low-severity stale-cache finding when stale versions exist', async () => {
    resetCounter();
    const result = await withHermeticHome(() => scan('/tmp/x', staleDiscovery));
    const f = result.findings.find(x => /stale plugin-cache/i.test(x.title || ''));
    assert.ok(f, 'expected a stale plugin-cache finding');
    assert.equal(f.severity, 'low');
    assert.equal(f.category, 'plugin-cache-hygiene');
  });

  it('states zero live-context impact and lists the stale keys', async () => {
    resetCounter();
    const result = await withHermeticHome(() => scan('/tmp/x', staleDiscovery));
    const f = result.findings.find(x => /stale plugin-cache/i.test(x.title || ''));
    assert.match(f.evidence, /zero live-context impact/i);
    assert.match(f.evidence, /mkt\/voyage\/5\.1\.1/);
    assert.match(f.evidence, /mkt\/config-audit\/5\.6\.0/);
  });

  it('does NOT fire when there are no stale versions', async () => {
    resetCounter();
    const result = await withHermeticHome(() => scan('/tmp/x', { files: [], staleCacheVersions: [] }));
    assert.ok(!result.findings.some(x => /stale plugin-cache/i.test(x.title || '')));
  });

  it('does NOT fire when discovery omits staleCacheVersions entirely', async () => {
    resetCounter();
    const result = await withHermeticHome(() => scan('/tmp/x', { files: [] }));
    assert.ok(!result.findings.some(x => /stale plugin-cache/i.test(x.title || '')));
  });
});