brain/tools/observer-transcript-parser.test.mjs

import { describe, it, expect } from 'vitest';
import { mkdtempSync, writeFileSync, rmSync } from 'node:fs';
import { tmpdir } from 'node:os';
import { join } from 'node:path';
import {
  parseTranscript,
  extractEnvironment,
  extractTaskSize,
  classifyPromptSignal,
  extractProcessEvents,
  parseRoutingTag,
  extractLastUserPromptText,
  classifyTask,
  extractTokenUsage,
  extractMcpServers,
  extractFileTypeDistribution,
  classifyFilePath,
} from './observer-transcript-parser.mjs';

describe('v4_signals + judge_spend_usd in episode', () => {
  it('episode carries v4_signals block (default zero/null when no runtime files)', () => {
    const ep = parseTranscript(JSON.stringify({
      sessionId: 'no-runtime-sess',
      message: { role: 'user', content: 'сделай фичу X' },
      timestamp: '2026-05-31T10:00:00.000Z',
    }), null, { routerStateBaseDir: tmpdir() });
    expect(ep.v4_signals).toEqual({
      rationalization_flag_count: 0, judge_verdict: null, safe_baseline_action: null, judge_calls: 0,
    });
    expect(typeof ep.task_cost.judge_spend_usd).toBe('number');
    expect(ep.schema_minor).toBe(4);
  });
});

// Build a JSONL transcript string from entry objects.
function jsonl(entries) {
  return entries.map((e) => JSON.stringify(e)).join('\n');
}

function userPrompt(text, ts, sessionId = 's1') {
  return { type: 'user', message: { role: 'user', content: text }, timestamp: ts, sessionId };
}
function assistantTurn(blocks, ts, sessionId = 's1') {
  return { type: 'assistant', message: { role: 'assistant', content: blocks }, timestamp: ts, sessionId };
}
function toolResult(toolUseId, isError, ts, sessionId = 's1') {
  return {
    type: 'user',
    message: {
      role: 'user',
      content: [{ type: 'tool_result', tool_use_id: toolUseId, content: 'r', is_error: isError }],
    },
    timestamp: ts,
    sessionId,
  };
}

describe('parseTranscript', () => {
  it('extracts task_id from sessionId in transcript', () => {
    const t = jsonl([userPrompt('add a feature', '2026-05-19T10:00:00Z', 'sess-xyz')]);
    expect(parseTranscript(t).task_id).toBe('sess-xyz');
  });

  it('falls back to provided sessionId when transcript has none', () => {
    const t = jsonl([
      { type: 'user', message: { role: 'user', content: 'hi' }, timestamp: '2026-05-19T10:00:00Z' },
    ]);
    expect(parseTranscript(t, 'fallback-id').task_id).toBe('fallback-id');
  });

  it('extracts timestamps from the last turn', () => {
    const t = jsonl([
      userPrompt('first', '2026-05-19T09:00:00Z'),
      userPrompt('second turn', '2026-05-19T10:00:00Z'),
      assistantTurn([{ type: 'text', text: 'done' }], '2026-05-19T10:05:00Z'),
    ]);
    const ep = parseTranscript(t);
    expect(ep.timestamps.started_at).toBe('2026-05-19T10:00:00Z');
    expect(ep.timestamps.ended_at).toBe('2026-05-19T10:05:00Z');
  });

  it('emits skill_invoked events for Skill tool_use', () => {
    const t = jsonl([
      userPrompt('do tdd', '2026-05-19T10:00:00Z'),
      assistantTurn(
        [{ type: 'tool_use', id: 't1', name: 'Skill', input: { skill: 'superpowers:test-driven-development' } }],
        '2026-05-19T10:01:00Z'
      ),
    ]);
    const skillEvents = parseTranscript(t).events.filter((e) => e.kind === 'skill_invoked');
    expect(skillEvents).toHaveLength(1);
    expect(skillEvents[0].skill).toBe('superpowers:test-driven-development');
  });

  it('emits a tool_summary event with per-tool counts', () => {
    const t = jsonl([
      userPrompt('work', '2026-05-19T10:00:00Z'),
      assistantTurn(
        [
          { type: 'tool_use', id: 't1', name: 'Read', input: {} },
          { type: 'tool_use', id: 't2', name: 'Read', input: {} },
          { type: 'tool_use', id: 't3', name: 'Bash', input: {} },
        ],
        '2026-05-19T10:01:00Z'
      ),
    ]);
    const summary = parseTranscript(t).events.find((e) => e.kind === 'tool_summary');
    expect(summary.counts).toEqual({ Read: 2, Bash: 1 });
  });

  it('node_chosen is the first skill invoked', () => {
    const t = jsonl([
      userPrompt('go', '2026-05-19T10:00:00Z'),
      assistantTurn(
        [{ type: 'tool_use', id: 't1', name: 'Skill', input: { skill: 'superpowers:brainstorming' } }],
        '2026-05-19T10:01:00Z'
      ),
    ]);
    expect(parseTranscript(t).primary_rationale.node_chosen).toBe('superpowers:brainstorming');
  });

  it('node_chosen is "direct" when no skill invoked', () => {
    const t = jsonl([
      userPrompt('go', '2026-05-19T10:00:00Z'),
      assistantTurn([{ type: 'tool_use', id: 't1', name: 'Read', input: {} }], '2026-05-19T10:01:00Z'),
    ]);
    expect(parseTranscript(t).primary_rationale.node_chosen).toBe('direct');
  });

  it('attaches chain_ref for a node that belongs to a chain', () => {
    const t = jsonl([
      userPrompt('go', '2026-05-19T10:00:00Z'),
      assistantTurn(
        [{ type: 'tool_use', id: 't1', name: 'Skill', input: { skill: 'billing-audit' } }],
        '2026-05-19T10:01:00Z'
      ),
    ]);
    expect(parseTranscript(t).primary_rationale.chain_ref).toEqual(['L13']);
  });

  it('sets chain_ref null for a direct episode', () => {
    const t = jsonl([
      userPrompt('go', '2026-05-19T10:00:00Z'),
      assistantTurn([{ type: 'tool_use', id: 't1', name: 'Read', input: {} }], '2026-05-19T10:01:00Z'),
    ]);
    expect(parseTranscript(t).primary_rationale.chain_ref).toBeNull();
  });

  it('hard_floor invoked when a superpowers skill is used', () => {
    const t = jsonl([
      userPrompt('go', '2026-05-19T10:00:00Z'),
      assistantTurn(
        [{ type: 'tool_use', id: 't1', name: 'Skill', input: { skill: 'superpowers:writing-plans' } }],
        '2026-05-19T10:01:00Z'
      ),
    ]);
    const hf = parseTranscript(t).primary_rationale.hard_floor;
    expect(hf.invoked).toBe(true);
    expect(hf.rules).toContain('Pravila §12');
  });

  it('hard_floor not invoked for a non-superpowers skill', () => {
    const t = jsonl([
      userPrompt('go', '2026-05-19T10:00:00Z'),
      assistantTurn(
        [{ type: 'tool_use', id: 't1', name: 'Skill', input: { skill: 'claude-md-management:claude-md-improver' } }],
        '2026-05-19T10:01:00Z'
      ),
    ]);
    expect(parseTranscript(t).primary_rationale.hard_floor.invoked).toBe(false);
  });

  it('classifies task from the user prompt text', () => {
    const cls = (text) =>
      parseTranscript(jsonl([userPrompt(text, '2026-05-19T10:00:00Z')])).primary_rationale.task_classification;
    expect(cls('почини баг в логине')).toBe('bugfix');
    expect(cls('добавь новую фичу экспорта')).toBe('feature');
    expect(cls('отрефактори этот модуль')).toBe('refactor');
    expect(cls('как это работает?')).toBe('question');
    expect(cls('обнови документацию readme')).toBe('docs');
  });

  it('scopes to the last turn — earlier turns are excluded', () => {
    const t = jsonl([
      userPrompt('turn 1', '2026-05-19T09:00:00Z'),
      assistantTurn(
        [{ type: 'tool_use', id: 't1', name: 'Skill', input: { skill: 'superpowers:brainstorming' } }],
        '2026-05-19T09:01:00Z'
      ),
      userPrompt('turn 2', '2026-05-19T10:00:00Z'),
      assistantTurn([{ type: 'tool_use', id: 't2', name: 'Read', input: {} }], '2026-05-19T10:01:00Z'),
    ]);
    const ep = parseTranscript(t);
    expect(ep.events.filter((e) => e.kind === 'skill_invoked')).toHaveLength(0);
    expect(ep.primary_rationale.node_chosen).toBe('direct');
  });

  it('does not treat a tool_result message as a turn boundary', () => {
    const t = jsonl([
      userPrompt('the only real prompt', '2026-05-19T10:00:00Z'),
      assistantTurn([{ type: 'tool_use', id: 't1', name: 'Bash', input: {} }], '2026-05-19T10:01:00Z'),
      toolResult('t1', false, '2026-05-19T10:02:00Z'),
      assistantTurn([{ type: 'tool_use', id: 't2', name: 'Edit', input: {} }], '2026-05-19T10:03:00Z'),
    ]);
    const summary = parseTranscript(t).events.find((e) => e.kind === 'tool_summary');
    expect(summary.counts).toEqual({ Bash: 1, Edit: 1 });
  });

  it('emits error events for tool_result with is_error', () => {
    const t = jsonl([
      userPrompt('go', '2026-05-19T10:00:00Z'),
      assistantTurn([{ type: 'tool_use', id: 't1', name: 'Bash', input: {} }], '2026-05-19T10:01:00Z'),
      toolResult('t1', true, '2026-05-19T10:02:00Z'),
    ]);
    const errs = parseTranscript(t).events.filter((e) => e.kind === 'error');
    expect(errs).toHaveLength(1);
    expect(errs[0].tool).toBe('Bash');
    expect(errs[0].summary).toBe('r');
  });

  it('skips broken JSONL lines without throwing', () => {
    const t = [
      JSON.stringify(userPrompt('go', '2026-05-19T10:00:00Z')),
      '{ broken json',
      JSON.stringify(assistantTurn([{ type: 'tool_use', id: 't1', name: 'Read', input: {} }], '2026-05-19T10:01:00Z')),
    ].join('\n');
    expect(() => parseTranscript(t)).not.toThrow();
    expect(parseTranscript(t).events.find((e) => e.kind === 'tool_summary').counts).toEqual({ Read: 1 });
  });

  it('path_type is regulated with a superpowers skill, improvised otherwise', () => {
    const reg = jsonl([
      userPrompt('go', '2026-05-19T10:00:00Z'),
      assistantTurn(
        [{ type: 'tool_use', id: 't1', name: 'Skill', input: { skill: 'superpowers:systematic-debugging' } }],
        '2026-05-19T10:01:00Z'
      ),
    ]);
    const imp = jsonl([
      userPrompt('go', '2026-05-19T10:00:00Z'),
      assistantTurn([{ type: 'tool_use', id: 't1', name: 'Read', input: {} }], '2026-05-19T10:01:00Z'),
    ]);
    expect(parseTranscript(reg).path_type).toBe('regulated');
    expect(parseTranscript(imp).path_type).toBe('improvised');
  });

  it('returns safe defaults for an empty transcript', () => {
    const ep = parseTranscript('');
    expect(ep.task_id).toBeTruthy();
    expect(ep.primary_rationale.node_chosen).toBe('direct');
    expect(ep.events).toEqual([]);
    expect(ep.outcome).toBe('unknown');
    expect(ep.schema_version).toBe(4);
    expect(ep.schema_minor).toBe(4);
  });

  it('produces a complete 7-field primary_rationale', () => {
    const ep = parseTranscript(jsonl([userPrompt('go', '2026-05-19T10:00:00Z')]));
    const r = ep.primary_rationale;
    for (const f of [
      'step',
      'node_chosen',
      'triggers_matched',
      'candidates_considered',
      'boundaries_applied',
      'hard_floor',
      'task_classification',
    ]) {
      expect(r[f]).toBeDefined();
    }
  });
});

describe('extractEnvironment', () => {
  it('reads economy_level from the ECONOMY MODE marker', () => {
    const entries = [
      userPrompt('=== ECONOMY MODE: 0% (пользователь указал явно) ===\nfix it', '2026-05-19T10:00:00Z'),
    ];
    expect(extractEnvironment(entries, 0).economy_level).toBe(0);
  });

  it('economy_level is null when no marker present', () => {
    const entries = [userPrompt('just do it', '2026-05-19T10:00:00Z')];
    expect(extractEnvironment(entries, 0).economy_level).toBeNull();
  });

  it('reads model from an assistant message', () => {
    const entries = [
      userPrompt('go', '2026-05-19T10:00:00Z'),
      { type: 'assistant', message: { role: 'assistant', model: 'claude-opus-4-7', content: [] }, timestamp: '2026-05-19T10:01:00Z', sessionId: 's1' },
    ];
    expect(extractEnvironment(entries, 0).model).toBe('claude-opus-4-7');
  });

  it('post_compaction is true when an isCompactSummary entry precedes the turn', () => {
    const entries = [
      { type: 'user', isCompactSummary: true, message: { role: 'user', content: 'summary' }, timestamp: '2026-05-19T09:00:00Z' },
      userPrompt('the real turn', '2026-05-19T10:00:00Z'),
    ];
    expect(extractEnvironment(entries, 1).post_compaction).toBe(true);
  });

  it('post_compaction is false with no compaction marker', () => {
    const entries = [userPrompt('turn one', '2026-05-19T09:00:00Z'), userPrompt('turn two', '2026-05-19T10:00:00Z')];
    expect(extractEnvironment(entries, 1).post_compaction).toBe(false);
  });

  it('session_turn counts real user prompts up to and including the turn start', () => {
    const entries = [
      userPrompt('one', '2026-05-19T09:00:00Z'),
      userPrompt('two', '2026-05-19T09:30:00Z'),
      userPrompt('three', '2026-05-19T10:00:00Z'),
    ];
    expect(extractEnvironment(entries, 2).session_turn).toBe(3);
  });

  it('session_turn counts only prompts after the last compaction', () => {
    const entries = [
      userPrompt('old 1', '2026-05-19T08:00:00Z'),
      userPrompt('old 2', '2026-05-19T08:30:00Z'),
      { type: 'user', isCompactSummary: true, message: { role: 'user', content: 'summary' }, timestamp: '2026-05-19T09:00:00Z' },
      userPrompt('after 1', '2026-05-19T09:30:00Z'),
      userPrompt('after 2 — turn', '2026-05-19T10:00:00Z'),
    ];
    expect(extractEnvironment(entries, 4).session_turn).toBe(2);
  });

  it('session_turn counts from the LAST compaction when several are present', () => {
    const entries = [
      { type: 'user', isCompactSummary: true, message: { role: 'user', content: 's1' }, timestamp: '2026-05-19T08:00:00Z' },
      userPrompt('mid 1', '2026-05-19T08:30:00Z'),
      { type: 'user', isCompactSummary: true, message: { role: 'user', content: 's2' }, timestamp: '2026-05-19T09:00:00Z' },
      userPrompt('after 1', '2026-05-19T09:30:00Z'),
      userPrompt('after 2 — turn', '2026-05-19T10:00:00Z'),
    ];
    expect(extractEnvironment(entries, 4).session_turn).toBe(2);
  });
});

describe('extractTaskSize', () => {
  it('counts tool calls and unique file paths', () => {
    const turn = [
      assistantTurn(
        [
          { type: 'tool_use', id: 't1', name: 'Read', input: { file_path: '/a.js' } },
          { type: 'tool_use', id: 't2', name: 'Edit', input: { file_path: '/a.js' } },
          { type: 'tool_use', id: 't3', name: 'Write', input: { file_path: '/b.js' } },
          { type: 'tool_use', id: 't4', name: 'Bash', input: {} },
        ],
        '2026-05-19T10:01:00Z'
      ),
    ];
    const size = extractTaskSize(turn);
    expect(size.tool_calls).toBe(4);
    expect(size.files_touched).toBe(2);
    expect(size.files.sort()).toEqual(['/a.js', '/b.js']);
  });

  it('returns zeros for an empty turn', () => {
    expect(extractTaskSize([])).toEqual({ tool_calls: 0, files_touched: 0, files: [] });
  });
});

describe('classifyPromptSignal', () => {
  it('detects corrections', () => {
    expect(classifyPromptSignal('не то, переделай')).toBe('correction');
    expect(classifyPromptSignal('почему ты это сделал')).toBe('correction');
  });
  it('detects widened correction phrases', () => {
    expect(classifyPromptSignal('не работает экспорт')).toBe('correction');
    expect(classifyPromptSignal('сломал тесты, верни как было')).toBe('correction');
    expect(classifyPromptSignal('опять не та колонка')).toBe('correction');
    expect(classifyPromptSignal('всё ещё падает')).toBe('correction');
    expect(classifyPromptSignal('откати последнюю правку')).toBe('correction');
    expect(classifyPromptSignal('this is still not working')).toBe('correction');
    expect(classifyPromptSignal('revert that change')).toBe('correction');
  });
  it('detects approvals', () => {
    expect(classifyPromptSignal('ок, спасибо')).toBe('approval');
    expect(classifyPromptSignal('готово, дальше')).toBe('approval');
  });
  it('detects a new task', () => {
    expect(classifyPromptSignal('добавь новую фичу экспорта в CSV')).toBe('new_task');
  });
  it('falls back to neutral', () => {
    expect(classifyPromptSignal('hmm')).toBe('neutral');
  });
});

describe('extractProcessEvents', () => {
  it('emits a hook_fired summary with per-hook counts and error count', () => {
    const turn = [
      { attachment: { type: 'hook_success', hookName: 'PreToolUse:Read' } },
      { attachment: { type: 'hook_success', hookName: 'PreToolUse:Read' } },
      { attachment: { type: 'hook_error', hookName: 'Stop:observer' } },
    ];
    const ev = extractProcessEvents(turn, 0, 0, 0).find((e) => e.kind === 'hook_fired');
    expect(ev.counts).toEqual({ 'PreToolUse:Read': 2, 'Stop:observer': 1 });
    expect(ev.errors).toBe(1);
  });

  it('emits an interrupt event for [Request interrupted by user]', () => {
    const turn = [
      { message: { role: 'user', content: [{ type: 'text', text: '[Request interrupted by user]' }] } },
    ];
    expect(extractProcessEvents(turn, 0, 0, 0).filter((e) => e.kind === 'interrupt')).toHaveLength(1);
  });

  it('emits a retry event when an errored tool is used again later', () => {
    const turn = [
      { message: { role: 'assistant', content: [{ type: 'tool_use', id: 'u1', name: 'Bash', input: {} }] } },
      { message: { role: 'user', content: [{ type: 'tool_result', tool_use_id: 'u1', is_error: true }] } },
      { message: { role: 'assistant', content: [{ type: 'tool_use', id: 'u2', name: 'Bash', input: {} }] } },
    ];
    expect(extractProcessEvents(turn, 0, 0, 0).filter((e) => e.kind === 'retry')).toHaveLength(1);
  });

  it('emits a time_burn event when the turn exceeds the threshold', () => {
    const ev = extractProcessEvents([], 0, 0, 1000000).find((e) => e.kind === 'time_burn');
    expect(ev.duration_ms).toBe(1000000);
  });

  it('emits a parse_gap event when the broken-line ratio is above threshold', () => {
    const ev = extractProcessEvents([], 3, 10, 0).find((e) => e.kind === 'parse_gap');
    expect(ev).toEqual({ kind: 'parse_gap', broken: 3, total: 10 });
  });

  it('emits nothing for a clean empty turn', () => {
    expect(extractProcessEvents([], 0, 0, 0)).toEqual([]);
  });

  it('emits unrecovered_error when the LAST tool_result in the turn is is_error', () => {
    const turn = [
      { message: { role: 'assistant', content: [{ type: 'tool_use', id: 'u1', name: 'Bash', input: {} }] } },
      { message: { role: 'user', content: [{ type: 'tool_result', tool_use_id: 'u1', is_error: true }] } },
    ];
    expect(extractProcessEvents(turn, 0, 0, 0).filter((e) => e.kind === 'unrecovered_error')).toHaveLength(1);
  });

  it('does NOT emit unrecovered_error when the turn ends on a successful tool_result', () => {
    const turn = [
      { message: { role: 'assistant', content: [{ type: 'tool_use', id: 'u1', name: 'Bash', input: {} }] } },
      { message: { role: 'user', content: [{ type: 'tool_result', tool_use_id: 'u1', is_error: true }] } },
      { message: { role: 'assistant', content: [{ type: 'tool_use', id: 'u2', name: 'Bash', input: {} }] } },
      { message: { role: 'user', content: [{ type: 'tool_result', tool_use_id: 'u2', is_error: false }] } },
    ];
    expect(extractProcessEvents(turn, 0, 0, 0).filter((e) => e.kind === 'unrecovered_error')).toHaveLength(0);
  });

  it('does NOT emit unrecovered_error for a turn with no tool_results at all', () => {
    const turn = [
      { message: { role: 'assistant', content: [{ type: 'text', text: 'just talking' }] } },
    ];
    expect(extractProcessEvents(turn, 0, 0, 0).filter((e) => e.kind === 'unrecovered_error')).toHaveLength(0);
  });
});

describe('parseRoutingTag', () => {
  it('parses a user_directed_method routing tag from assistant text', () => {
    const turn = [
      assistantTurn(
        [{ type: 'text', text: 'ok\n<!-- routing: provenance=user_directed_method node=discovery-interview counterfactual=brainstorming -->' }],
        '2026-05-19T10:01:00Z'
      ),
    ];
    expect(parseRoutingTag(turn)).toEqual({
      kind: 'user_directed_method',
      node: 'discovery-interview',
      claude_would_have_chosen: 'brainstorming',
    });
  });

  it('returns null when no tag is present', () => {
    const turn = [assistantTurn([{ type: 'text', text: 'plain answer' }], '2026-05-19T10:01:00Z')];
    expect(parseRoutingTag(turn)).toBeNull();
  });
});

describe('parseTranscript — v4 episode (Phase 2 Task 15 bump)', () => {
  it('produces schema_version 4 and all v2+ fields', () => {
    const t = jsonl([
      userPrompt('=== ECONOMY MODE: 0% ===\nдобавь фичу', '2026-05-19T10:00:00Z', 'sess-v2'),
      assistantTurn([{ type: 'tool_use', id: 't1', name: 'Read', input: { file_path: '/x.js' } }], '2026-05-19T10:01:00Z', 'sess-v2'),
    ]);
    const ep = parseTranscript(t);
    expect(ep.schema_version).toBe(4);
    expect(ep.schema_minor).toBe(4);
    expect('classifier_output' in ep).toBe(true);
    expect('degraded_mode' in ep).toBe(true);
    expect('classifier_model' in ep.environment).toBe(true);
    expect(ep.task_ref).toBe('sess-v2');
    expect(ep.outcome).toBe('unknown');
    expect(ep.prompt_signal).toBe('new_task');
    expect(ep.decision_provenance).toEqual({ kind: 'autonomous', claude_would_have_chosen: null });
    expect(ep.environment.economy_level).toBe(0);
    expect(ep.task_size).toEqual({ tool_calls: 1, files_touched: 1, files: ['/x.js'] });
  });

  it('records decision_provenance from a routing tag', () => {
    const t = jsonl([
      userPrompt('запусти discovery-interview', '2026-05-19T10:00:00Z', 'sess-tag'),
      assistantTurn(
        [{ type: 'text', text: '<!-- routing: provenance=user_directed_method node=discovery-interview counterfactual=brainstorming -->' }],
        '2026-05-19T10:01:00Z',
        'sess-tag'
      ),
    ]);
    const ep = parseTranscript(t);
    expect(ep.decision_provenance.kind).toBe('user_directed_method');
    expect(ep.decision_provenance.claude_would_have_chosen).toBe('brainstorming');
  });
});

describe('extractLastUserPromptText', () => {
  it('returns the text of the last real user prompt', () => {
    const t = jsonl([
      userPrompt('first turn', '2026-05-19T09:00:00Z'),
      userPrompt('second and last', '2026-05-19T10:00:00Z'),
    ]);
    expect(extractLastUserPromptText(t)).toBe('second and last');
  });
});

describe('parseTranscript — user_chose_from_options', () => {
  it('classifies as user_chose_from_options when last assistant offered options and user picked one', () => {
    const lines = [
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:00.000Z',
        sessionId: 's1',
        message: { role: 'user', content: 'continue task' },
      }),
      JSON.stringify({
        type: 'assistant',
        timestamp: '2026-05-19T10:00:01.000Z',
        sessionId: 's1',
        message: {
          role: 'assistant',
          content: [{ type: 'text', text: 'Choose:\n1. First option\n2. Second option\n3. Third option' }],
        },
      }),
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:10.000Z',
        sessionId: 's1',
        message: { role: 'user', content: '2 делаем' },
      }),
      JSON.stringify({
        type: 'assistant',
        timestamp: '2026-05-19T10:00:11.000Z',
        sessionId: 's1',
        message: { role: 'assistant', content: [{ type: 'text', text: 'ok' }] },
      }),
    ].join('\n');
    const ep = parseTranscript(lines, 'fallback');
    expect(ep.decision_provenance).toEqual({
      kind: 'user_chose_from_options',
      node: 'Second option',
      options_offered: ['First option', 'Second option', 'Third option'],
      claude_would_have_chosen: 'First option',
    });
  });

  it('falls back to autonomous when last assistant had no options', () => {
    const lines = [
      JSON.stringify({
        type: 'assistant',
        timestamp: '2026-05-19T10:00:00.000Z',
        sessionId: 's1',
        message: { role: 'assistant', content: [{ type: 'text', text: 'just a paragraph' }] },
      }),
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:10.000Z',
        sessionId: 's1',
        message: { role: 'user', content: '2 делаем' },
      }),
      JSON.stringify({
        type: 'assistant',
        timestamp: '2026-05-19T10:00:11.000Z',
        sessionId: 's1',
        message: { role: 'assistant', content: [{ type: 'text', text: 'ok' }] },
      }),
    ].join('\n');
    const ep = parseTranscript(lines, 'fallback');
    expect(ep.decision_provenance).toEqual({ kind: 'autonomous', claude_would_have_chosen: null });
  });

  it('routing-tag user_directed_method preserved when no choice detected', () => {
    const lines = [
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:00.000Z',
        sessionId: 's1',
        message: { role: 'user', content: 'запусти brainstorming' },
      }),
      JSON.stringify({
        type: 'assistant',
        timestamp: '2026-05-19T10:00:01.000Z',
        sessionId: 's1',
        message: {
          role: 'assistant',
          content: [{ type: 'text', text: 'ok\n<!-- routing: provenance=user_directed_method node=brainstorming counterfactual=writing-plans -->' }],
        },
      }),
    ].join('\n');
    const ep = parseTranscript(lines, 'fallback');
    expect(ep.decision_provenance.kind).toBe('user_directed_method');
  });
});

describe('parseTranscript — synthetic user messages skipped', () => {
  it('captures economy_level from genuine prompt even when skill-content message follows', () => {
    const lines = [
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:00.000Z',
        sessionId: 's1',
        message: { role: 'user', content: 'почини баг в парсере' },
      }),
      JSON.stringify({
        type: 'attachment',
        timestamp: '2026-05-19T10:00:00.500Z',
        sessionId: 's1',
        attachment: {
          type: 'hook_additional_context',
          hookName: 'UserPromptSubmit',
          content: ['=== ECONOMY MODE: 5% (тест) ===\nинструкции режима...'],
        },
      }),
      JSON.stringify({
        type: 'assistant',
        timestamp: '2026-05-19T10:00:01.000Z',
        sessionId: 's1',
        message: { role: 'assistant', content: [{ type: 'text', text: 'смотрю' }] },
      }),
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:02.000Z',
        sessionId: 's1',
        message: { role: 'user', content: 'Base directory for this skill: C:\\path\\skill\n\n# Some Skill\nbody' },
      }),
      JSON.stringify({
        type: 'assistant',
        timestamp: '2026-05-19T10:00:03.000Z',
        sessionId: 's1',
        message: { role: 'assistant', content: [{ type: 'text', text: 'готово' }] },
      }),
    ].join('\n');
    const ep = parseTranscript(lines, 'fallback');
    expect(ep.environment.economy_level).toBe(5);
    expect(ep.primary_rationale.task_classification).toBe('bugfix');
  });

  it('extractLastUserPromptText skips skill-content and returns genuine prompt', () => {
    const lines = [
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:00.000Z',
        sessionId: 's1',
        message: { role: 'user', content: 'добавь колонку Город' },
      }),
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:02.000Z',
        sessionId: 's1',
        message: { role: 'user', content: 'Base directory for this skill: C:\\x\n# Skill body' },
      }),
    ].join('\n');
    expect(extractLastUserPromptText(lines)).toBe('добавь колонку Город');
  });

  it('extractLastUserPromptText skips local-command output and interrupt markers', () => {
    const lines = [
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:00.000Z',
        sessionId: 's1',
        message: { role: 'user', content: 'сделай X' },
      }),
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:01.000Z',
        sessionId: 's1',
        message: { role: 'user', content: '[Request interrupted by user]' },
      }),
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:02.000Z',
        sessionId: 's1',
        message: { role: 'user', content: '<local-command-stdout>some output</local-command-stdout>' },
      }),
    ].join('\n');
    expect(extractLastUserPromptText(lines)).toBe('сделай X');
  });
});

describe('parseTranscript — AskUserQuestion in-turn choice', () => {
  it('classifies user_chose_from_options when an AskUserQuestion option was clicked', () => {
    const lines = [
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:00.000Z',
        sessionId: 's1',
        message: { role: 'user', content: 'построй фичу' },
      }),
      JSON.stringify({
        type: 'assistant',
        timestamp: '2026-05-19T10:00:01.000Z',
        sessionId: 's1',
        message: {
          role: 'assistant',
          content: [{ type: 'tool_use', name: 'AskUserQuestion', input: {} }],
        },
      }),
      JSON.stringify({
        type: 'user',
        timestamp: '2026-05-19T10:00:05.000Z',
        sessionId: 's1',
        message: { role: 'user', content: [{ type: 'tool_result', content: 'User has answered...' }] },
        toolUseResult: {
          questions: [
            {
              question: 'Каким режимом?',
              options: [{ label: 'Subagent-Driven' }, { label: 'Inline execution' }],
            },
          ],
          answers: { 'Каким режимом?': 'Inline execution' },
        },
      }),
      JSON.stringify({
        type: 'assistant',
        timestamp: '2026-05-19T10:00:06.000Z',
        sessionId: 's1',
        message: { role: 'assistant', content: [{ type: 'text', text: 'ok' }] },
      }),
    ].join('\n');
    const ep = parseTranscript(lines, 'fallback');
    expect(ep.decision_provenance).toEqual({
      kind: 'user_chose_from_options',
      node: 'Inline execution',
      options_offered: ['Subagent-Driven', 'Inline execution'],
      claude_would_have_chosen: 'Subagent-Driven',
    });
  });
});

describe('parseTranscript — uuid-dedup for duplicated snapshots (quirk #101)', () => {
  // Claude Code's transcript file accumulates duplicated context-rebuild
  // snapshots; the same entry is re-printed with the SAME `uuid`. Without
  // dedup, session_turn / task_size / events double-count and become
  // non-monotonic across episodes parsed at different file-growth states.
  // Root fix: dedup by uuid in parseLines.

  it('collapses duplicated-uuid user prompts (session_turn counts once)', () => {
    const dup = 'aaaa-bbbb-cccc';
    const lines = [
      JSON.stringify({
        uuid: dup,
        type: 'user',
        message: { role: 'user', content: 'hi' },
        timestamp: '2026-05-19T10:00:00Z',
        sessionId: 's1',
      }),
      JSON.stringify({
        uuid: dup,
        type: 'user',
        message: { role: 'user', content: 'hi' },
        timestamp: '2026-05-19T10:00:00Z',
        sessionId: 's1',
      }),
    ].join('\n');
    const ep = parseTranscript(lines, 'fallback');
    expect(ep.environment.session_turn).toBe(1);
  });

  it('preserves distinct-uuid entries (no over-dedup)', () => {
    const lines = [
      JSON.stringify({
        uuid: 'one',
        type: 'user',
        message: { role: 'user', content: 'first' },
        timestamp: '2026-05-19T09:00:00Z',
        sessionId: 's1',
      }),
      JSON.stringify({
        uuid: 'two',
        type: 'user',
        message: { role: 'user', content: 'second' },
        timestamp: '2026-05-19T10:00:00Z',
        sessionId: 's1',
      }),
    ].join('\n');
    const ep = parseTranscript(lines, 'fallback');
    expect(ep.environment.session_turn).toBe(2);
  });

  it('entries without uuid pass through unchanged (synthetic fixtures unaffected)', () => {
    // Existing tests build entries without `uuid` — must still work.
    const t = jsonl([
      userPrompt('one', '2026-05-19T09:00:00Z'),
      userPrompt('two', '2026-05-19T10:00:00Z'),
    ]);
    expect(parseTranscript(t).environment.session_turn).toBe(2);
  });

  it('collapses duplicated-uuid assistant turns (task_size counted once)', () => {
    const lines = [
      JSON.stringify({
        uuid: 'u-prompt',
        type: 'user',
        message: { role: 'user', content: 'go' },
        timestamp: '2026-05-19T10:00:00Z',
        sessionId: 's1',
      }),
      JSON.stringify({
        uuid: 'u-asst',
        type: 'assistant',
        message: {
          role: 'assistant',
          content: [
            { type: 'tool_use', id: 't1', name: 'Read', input: { file_path: '/x.js' } },
            { type: 'tool_use', id: 't2', name: 'Edit', input: { file_path: '/x.js' } },
          ],
        },
        timestamp: '2026-05-19T10:01:00Z',
        sessionId: 's1',
      }),
      // EXACT duplicate of the assistant entry — snapshot rebuild artifact
      JSON.stringify({
        uuid: 'u-asst',
        type: 'assistant',
        message: {
          role: 'assistant',
          content: [
            { type: 'tool_use', id: 't1', name: 'Read', input: { file_path: '/x.js' } },
            { type: 'tool_use', id: 't2', name: 'Edit', input: { file_path: '/x.js' } },
          ],
        },
        timestamp: '2026-05-19T10:01:00Z',
        sessionId: 's1',
      }),
    ].join('\n');
    const ep = parseTranscript(lines, 'fallback');
    expect(ep.task_size.tool_calls).toBe(2);
    expect(ep.task_size.files_touched).toBe(1);
  });
});

describe('extractEnvironment — parallel_session narrowed to tool_result evidence', () => {
  // Collision detection scans ONLY tool_result content (real command output).
  // Prose mentions in user prompts / assistant text — including analysis text
  // that references collision phrases — must not trigger.
  const userMsg = (text) => ({
    message: { role: 'user', content: text },
    timestamp: '2026-05-19T10:00:00.000Z',
  });
  const assistantText = (text) => ({
    message: { role: 'assistant', content: [{ type: 'text', text }] },
    timestamp: '2026-05-19T10:00:01.000Z',
  });
  const toolResultText = (text, isError = false) => ({
    message: {
      role: 'user',
      content: [{ type: 'tool_result', tool_use_id: 't1', content: text, is_error: isError }],
    },
    timestamp: '2026-05-19T10:00:02.000Z',
  });

  it('is false when a USER prompt only mentions a collision marker', () => {
    const env = extractEnvironment([userMsg('в моём ходе появился чужой staged-файл — что делать?')], 0);
    expect(env.parallel_session).toBe(false);
  });

  it('is false when ASSISTANT prose mentions a collision marker (e.g., analysis text)', () => {
    const env = extractEnvironment(
      [userMsg('go'), assistantText('я писал про чужой staged и foreign git index в разборе')],
      0
    );
    expect(env.parallel_session).toBe(false);
  });

  it('is true when a tool_result (Bash output) contains "foreign git index"', () => {
    const env = extractEnvironment(
      [
        userMsg('go'),
        toolResultText('fatal: another git process is running\nforeign git index detected', true),
      ],
      0
    );
    expect(env.parallel_session).toBe(true);
  });

  it('is true when a tool_result contains "чужой staged"', () => {
    const env = extractEnvironment(
      [userMsg('go'), toolResultText('error: чужой staged-файл присутствует в индексе')],
      0
    );
    expect(env.parallel_session).toBe(true);
  });

  it('is false when there are no tool_result entries at all', () => {
    const env = extractEnvironment([userMsg('говорим о параллельных сессиях вообще')], 0);
    expect(env.parallel_session).toBe(false);
  });

  it('handles tool_result with content as a structured array of text blocks', () => {
    const env = extractEnvironment(
      [
        userMsg('go'),
        {
          message: {
            role: 'user',
            content: [
              {
                type: 'tool_result',
                tool_use_id: 't1',
                content: [{ type: 'text', text: 'index.lock exists, another process is holding it' }],
              },
            ],
          },
          timestamp: '2026-05-19T10:00:02.000Z',
        },
      ],
      0
    );
    expect(env.parallel_session).toBe(true);
  });
});

describe('classifyTask — extended dictionary (Task 1)', () => {
  it('classifies analysis prompts', () => {
    expect(classifyTask('проанализируй данные наблюдателя')).toBe('analysis');
    expect(classifyTask('review the recent changes')).toBe('analysis');
    expect(classifyTask('посмотри что в логах')).toBe('analysis');
  });
  it('classifies memory-sync prompts', () => {
    expect(classifyTask('обнови эталон проекта')).toBe('memory-sync');
    expect(classifyTask('sync memory с последним push')).toBe('memory-sync');
    expect(classifyTask('обнови MEMORY.md')).toBe('memory-sync');
  });
  it('classifies regulatory-bump prompts', () => {
    expect(classifyTask('обнови CLAUDE.md после правки')).toBe('regulatory-bump');
    expect(classifyTask('правка Pravila §16')).toBe('regulatory-bump');
    expect(classifyTask('обнови PSR_v1 v3.18')).toBe('regulatory-bump');
  });
  it('classifies release prompts', () => {
    expect(classifyTask('push origin main')).toBe('release');
    expect(classifyTask('merge feature-branch')).toBe('release');
    expect(classifyTask('сделай release v2')).toBe('release');
  });
  it('does NOT classify a plain commit (no push/release) as release', () => {
    expect(classifyTask('commit the auth changes')).not.toBe('release');
  });
  it('classifies cleanup prompts', () => {
    expect(classifyTask('убери временные файлы')).toBe('cleanup');
    expect(classifyTask('cleanup тулчейн')).toBe('cleanup');
  });
  it('classifies monitoring prompts', () => {
    expect(classifyTask('проверь состояние портала')).toBe('monitoring');
    expect(classifyTask('status of CI')).toBe('monitoring');
  });
  it('classifies planning prompts', () => {
    expect(classifyTask('план рефакторинга биллинга')).toBe('planning');
    expect(classifyTask('design новую фичу')).toBe('planning');
    expect(classifyTask('обсудим архитектуру')).toBe('planning');
  });
  it('preserves existing classes — refactor stays refactor', () => {
    expect(classifyTask('рефактор биллинг-сервиса')).toBe('refactor');
  });
  it('preserves existing classes — bugfix stays bugfix', () => {
    expect(classifyTask('почини баг в logger')).toBe('bugfix');
  });
});

describe('extractTokenUsage (Task 2)', () => {
  it('sums input/output/cache fields across multiple assistant messages', () => {
    const turn = [
      { message: { usage: { input_tokens: 10, output_tokens: 5, cache_read_input_tokens: 100, cache_creation_input_tokens: 50 } } },
      { message: { usage: { input_tokens: 8, output_tokens: 3, cache_read_input_tokens: 80, cache_creation_input_tokens: 20 } } },
    ];
    expect(extractTokenUsage(turn)).toEqual({
      input_tokens: 18, output_tokens: 8, cache_read_input_tokens: 180,
      cache_creation_input_tokens: 70, web_search_requests: 0, web_fetch_requests: 0, iterations: 0,
      classifier_input_tokens: 0, classifier_output_tokens: 0,
      self_assessment_input_tokens: 0, self_assessment_output_tokens: 0,
      reviewer_input_tokens: 0, reviewer_output_tokens: 0,
      reviewer_subagent_usd: 0, reviewer_direct_fallback_usd: 0,
    });
  });
  it('captures server_tool_use bonus fields (web_search/web_fetch)', () => {
    const turn = [
      { message: { usage: { input_tokens: 5, output_tokens: 2, server_tool_use: { web_search_requests: 3, web_fetch_requests: 1 } } } },
    ];
    const result = extractTokenUsage(turn);
    expect(result.web_search_requests).toBe(3);
    expect(result.web_fetch_requests).toBe(1);
  });
  it('captures iterations (extended-thinking detector)', () => {
    const turn = [
      { message: { usage: { input_tokens: 100, output_tokens: 50, iterations: 4 } } },
    ];
    expect(extractTokenUsage(turn).iterations).toBe(4);
  });
  it('returns zero-filled object when no usage present', () => {
    const turn = [
      { message: {} },
      { message: { usage: null } },
    ];
    expect(extractTokenUsage(turn)).toEqual({
      input_tokens: 0, output_tokens: 0, cache_read_input_tokens: 0,
      cache_creation_input_tokens: 0, web_search_requests: 0, web_fetch_requests: 0, iterations: 0,
      classifier_input_tokens: 0, classifier_output_tokens: 0,
      self_assessment_input_tokens: 0, self_assessment_output_tokens: 0,
      reviewer_input_tokens: 0, reviewer_output_tokens: 0,
      reviewer_subagent_usd: 0, reviewer_direct_fallback_usd: 0,
    });
  });
  it('handles empty/null turn safely', () => {
    const zeroShape = {
      input_tokens: 0, output_tokens: 0, cache_read_input_tokens: 0,
      cache_creation_input_tokens: 0, web_search_requests: 0, web_fetch_requests: 0, iterations: 0,
      classifier_input_tokens: 0, classifier_output_tokens: 0,
      self_assessment_input_tokens: 0, self_assessment_output_tokens: 0,
      reviewer_input_tokens: 0, reviewer_output_tokens: 0,
      reviewer_subagent_usd: 0, reviewer_direct_fallback_usd: 0,
    };
    expect(extractTokenUsage([])).toEqual(zeroShape);
    expect(extractTokenUsage(null)).toEqual(zeroShape);
  });
  it('safely skips entries where usage is a non-object primitive (defensive guard)', () => {
    const turn = [
      { message: { usage: 42 } }, // malformed — usage as primitive
      { message: { usage: { input_tokens: 5, output_tokens: 3 } } },
    ];
    const r = extractTokenUsage(turn);
    expect(r.input_tokens).toBe(5);
    expect(r.output_tokens).toBe(3);
  });
});

describe('parseTranscript — task_cost integration (Task 2)', () => {
  it('attaches task_cost to a v2 episode', () => {
    const lines = [
      JSON.stringify({ type: 'user', message: { role: 'user', content: [{ type: 'text', text: 'implement feature X' }] } }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [{ type: 'text', text: 'done' }], usage: { input_tokens: 42, output_tokens: 7 } } }),
    ];
    const result = parseTranscript(lines.join('\n'));
    expect(result).not.toBeNull();
    expect(result.task_cost).toBeDefined();
    expect(result.task_cost.input_tokens).toBe(42);
    expect(result.task_cost.output_tokens).toBe(7);
  });
  it('attaches zero-filled task_cost when no usage in transcript', () => {
    const lines = [
      JSON.stringify({ type: 'user', message: { role: 'user', content: [{ type: 'text', text: 'do something' }] } }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [{ type: 'text', text: 'ok' }] } }),
    ];
    const result = parseTranscript(lines.join('\n'));
    expect(result).not.toBeNull();
    expect(result.task_cost).toEqual({
      input_tokens: 0, output_tokens: 0, cache_read_input_tokens: 0,
      cache_creation_input_tokens: 0, web_search_requests: 0, web_fetch_requests: 0, iterations: 0,
      classifier_input_tokens: 0, classifier_output_tokens: 0,
      self_assessment_input_tokens: 0, self_assessment_output_tokens: 0,
      reviewer_input_tokens: 0, reviewer_output_tokens: 0,
      reviewer_subagent_usd: 0, reviewer_direct_fallback_usd: 0,
      judge_spend_usd: 0,
    });
  });
});

import { extractAskUserQuestionEvents } from './observer-transcript-parser.mjs';

describe('extractAskUserQuestionEvents (Task 4)', () => {
  it('records "option" when user picked an offered label exactly', () => {
    const turn = [
      { message: { content: [{ type: 'tool_use', name: 'AskUserQuestion',
        input: { questions: [{ question: 'q1', options: [{ label: 'A' }, { label: 'B' }] }] } }] } },
      { toolUseResult: { questions: [{ question: 'q1', options: [{ label: 'A' }, { label: 'B' }] }],
        answers: { q1: 'A' } } },
    ];
    const evs = extractAskUserQuestionEvents(turn);
    expect(evs).toHaveLength(1);
    expect(evs[0].kind).toBe('ask_user_question');
    expect(evs[0].question_count).toBe(1);
    expect(evs[0].answer_kind).toBe('option');
  });
  it('records "custom" when user wrote free-text Other', () => {
    const turn = [
      { message: { content: [{ type: 'tool_use', name: 'AskUserQuestion',
        input: { questions: [{ question: 'q1', options: [{ label: 'A' }, { label: 'B' }] }] } }] } },
      { toolUseResult: { questions: [{ question: 'q1', options: [{ label: 'A' }, { label: 'B' }] }],
        answers: { q1: 'C — мой вариант' } } },
    ];
    const evs = extractAskUserQuestionEvents(turn);
    expect(evs[0].answer_kind).toBe('custom');
  });
  it('records "no_answer" when answer key missing', () => {
    const turn = [
      { toolUseResult: { questions: [{ question: 'q1', options: [{ label: 'A' }, { label: 'B' }] }],
        answers: {} } },
    ];
    const evs = extractAskUserQuestionEvents(turn);
    expect(evs[0].answer_kind).toBe('no_answer');
  });
  it('reports question_count from toolUseResult.questions length', () => {
    const turn = [
      { toolUseResult: { questions: [
          { question: 'q1', options: [{ label: 'A' }] },
          { question: 'q2', options: [{ label: 'B' }] },
        ], answers: { q1: 'A', q2: 'B' } } },
    ];
    const evs = extractAskUserQuestionEvents(turn);
    expect(evs).toHaveLength(2);   // one event per question
    expect(evs[0].question_count).toBe(2);
    expect(evs[1].question_count).toBe(2);
  });
  it('returns empty for a turn without AskUserQuestion', () => {
    expect(extractAskUserQuestionEvents([{ message: { content: [{ type: 'tool_use', name: 'Read' }] } }])).toEqual([]);
  });
  it('handles null/undefined turn safely', () => {
    expect(extractAskUserQuestionEvents(null)).toEqual([]);
    expect(extractAskUserQuestionEvents(undefined)).toEqual([]);
    expect(extractAskUserQuestionEvents([])).toEqual([]);
  });
});

describe('parseTranscript — ask_user_question events (Task 4)', () => {
  it('emits ask_user_question event from AskUserQuestion + toolUseResult', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's1' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'делай' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [
        { type: 'tool_use', id: 't1', name: 'AskUserQuestion',
          input: { questions: [{ question: 'q', options: [{ label: 'X' }, { label: 'Y' }] }] } },
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:00:01Z' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: [
        { type: 'tool_result', tool_use_id: 't1' },
      ] }, toolUseResult: { questions: [{ question: 'q', options: [{ label: 'X' }, { label: 'Y' }] }],
        answers: { q: 'X' } }, uuid: 'u3', timestamp: '2026-05-20T00:00:02Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    const aq = ep.events.filter((e) => e.kind === 'ask_user_question');
    expect(aq).toHaveLength(1);
    expect(aq[0].answer_kind).toBe('option');
  });
});

import {
  extractTriggers,
  extractCandidates,
  extractBoundaries,
} from './observer-transcript-parser.mjs';

describe('reasoning capture heuristics (Task 6)', () => {
  const mkTurn = (txt) => [{ message: { role: 'assistant', content: [{ type: 'text', text: txt }] } }];

  describe('extractTriggers', () => {
    it('finds Pravila §N references', () => {
      expect(extractTriggers(mkTurn('per Pravila §12.2 hard-rule'))).toContain('Pravila §12.2');
    });
    it('finds ADR references', () => {
      expect(extractTriggers(mkTurn('see ADR-011 anchor'))).toContain('ADR-011');
    });
    it('finds PSR_v1 R refs', () => {
      expect(extractTriggers(mkTurn('PSR_v1 R10.1 requires it'))).toContain('PSR_v1 R10.1');
    });
    it('finds routing-off-phase L refs from canonical form', () => {
      expect(extractTriggers(mkTurn('routing-off-phase L12 chain'))).toContain('routing-off-phase L12');
    });
    it('finds hard-rule / hard-floor (case-insensitive)', () => {
      const res = extractTriggers(mkTurn('this is a hard-rule per §15'));
      expect(res.some((t) => t.toLowerCase().includes('hard-rule'))).toBe(true);
    });
    it('deduplicates repeated triggers', () => {
      const res = extractTriggers(mkTurn('Pravila §16 and Pravila §16 again'));
      expect(res.filter((t) => t === 'Pravila §16')).toHaveLength(1);
    });
    it('returns empty for plain prose', () => {
      expect(extractTriggers(mkTurn('just plain text'))).toEqual([]);
    });
    it('safe on null/empty', () => {
      expect(extractTriggers(null)).toEqual([]);
      expect(extractTriggers([])).toEqual([]);
    });
  });

  describe('extractCandidates', () => {
    // Only items that look like router-node identifiers are accepted:
    // - a known node from tools/observer-known-nodes.txt
    // - a key from tools/observer-chain-map.json (e.g. superpowers:brainstorming)
    // - a tooling ID matching ^#\d+$ (Прил. Н)
    // - the sentinel "direct"
    // Free-form prose bullets / numbered procedure steps / code snippets are rejected.
    it('extracts numbered options that are known node names', () => {
      const c = extractCandidates(
        mkTurn('1. brainstorming\n2. subagent-driven-development\n3. direct')
      );
      expect(c).toEqual(['brainstorming', 'subagent-driven-development', 'direct']);
    });
    it('accepts namespaced plugin:skill form from the chain map', () => {
      const c = extractCandidates(
        mkTurn('1. superpowers:brainstorming\n2. claude-md-management:claude-md-improver')
      );
      expect(c).toEqual(['superpowers:brainstorming', 'claude-md-management:claude-md-improver']);
    });
    it('accepts tooling IDs like #25 / #74', () => {
      expect(extractCandidates(mkTurn('1. #25\n2. #74'))).toEqual(['#25', '#74']);
    });
    it('strips simple markdown wrappers before checking', () => {
      const c = extractCandidates(
        mkTurn('1. **brainstorming**\n2. `writing-plans`\n3. discovery-interview')
      );
      expect(c).toEqual(['brainstorming', 'writing-plans', 'discovery-interview']);
    });
    it('drops free-form prose bullets even when ≥2 are present', () => {
      // Repro from docs/observer/episodes-2026-05.jsonl: analysis bullets with
      // bold-prefix sentence text were going straight into candidates_considered.
      const text =
        '- **Hard-floor работает только для §12 Superpowers** (14 раз). §14/§15 в журнале не оставили следов.\n' +
        '- **На feature/planning я не ищу триггеры** — 0% матча.\n' +
        '- **Метка `regulated` врёт** в 79% случаев — нет настоящего применения границ.';
      expect(extractCandidates(mkTurn(text))).toEqual([]);
    });
    it('drops numbered procedure-step text (real episode repro)', () => {
      const text =
        '1. **Hard-floor check** — Pravila §12 (Superpowers) / §14 (Queen) / §15.\n' +
        '2. **Классификация** — определяю тип задачи (feature/bugfix/planning).\n' +
        '3. **Trigger-based node selection** — по реестру Tooling Прил. Н §4.X.\n' +
        '4. **Canonical chain check** — смотрю L1-L15.\n' +
        '5. **Execution** — иду делать.';
      expect(extractCandidates(mkTurn(text))).toEqual([]);
    });
    it('drops code-snippet bullets (regex patterns etc.)', () => {
      const text =
        '- regex `(?:^|[\\s\\"\\\'])(tools\\/[\\w-]+\\.(?:mjs|py|sh))` → имя файла.\n' +
        '- fallback `inline:<sha256(command).slice(0,16)>` — стабильно.';
      expect(extractCandidates(mkTurn(text))).toEqual([]);
    });
    it('filters a mixed list down to just the real nodes', () => {
      const text =
        '1. brainstorming\n' +
        '2. resolver + tests\n' +
        '3. discovery-interview\n' +
        '4. parser extension + tests + smoke';
      expect(extractCandidates(mkTurn(text))).toEqual(['brainstorming', 'discovery-interview']);
    });
    it('returns empty when only one real node survives the filter', () => {
      // ≥2 raw items but only 1 known-node → not enough signal, drop.
      expect(extractCandidates(mkTurn('1. brainstorming\n2. некий текст'))).toEqual([]);
    });
    it('prefers numbered over bullets when both lists contain known nodes', () => {
      const c = extractCandidates(
        mkTurn('1. brainstorming\n2. writing-plans\n- discovery-interview\n- regression')
      );
      expect(c).toEqual(['brainstorming', 'writing-plans']);
    });
    it('returns empty for prose', () => {
      expect(extractCandidates(mkTurn('просто текст'))).toEqual([]);
    });
    it('safe on null/empty', () => {
      expect(extractCandidates(null)).toEqual([]);
    });
  });

  describe('extractBoundaries', () => {
    it('finds ADR + PSR + Pravila refs', () => {
      const b = extractBoundaries(mkTurn('per ADR-011 + PSR_v1 R16 + Pravila §16.2'));
      expect(b).toContain('ADR-011');
      expect(b.some((x) => x.includes('PSR_v1 R16'))).toBe(true);
      expect(b).toContain('Pravila §16.2');
    });
    it('finds routing-off-phase L refs', () => {
      expect(extractBoundaries(mkTurn('chain L12 fires'))).toEqual(expect.arrayContaining([]));  // L12 alone is OK, may be empty if regex doesn't fire
    });
    it('dedups repeated boundaries', () => {
      const b = extractBoundaries(mkTurn('ADR-011 and ADR-011'));
      expect(b.filter((x) => x === 'ADR-011')).toHaveLength(1);
    });
    it('safe on null/empty', () => {
      expect(extractBoundaries(null)).toEqual([]);
    });
  });
});

describe('parseTranscript — heuristic primary_rationale (Task 6)', () => {
  it('populates triggers_matched / candidates_considered / boundaries_applied', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's1' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'делай' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [
        { type: 'text', text: 'per Pravila §12.2 hard-rule\n1. brainstorming\n2. direct\nADR-011 applies.' }
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:01:00Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    expect(ep.primary_rationale.triggers_matched).toContain('Pravila §12.2');
    expect(ep.primary_rationale.candidates_considered).toContain('brainstorming');
    expect(ep.primary_rationale.boundaries_applied).toContain('ADR-011');
  });
});

describe('error event differentiation (Task 7)', () => {
  it('attributes error to the failing tool and captures summary', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'делай' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [
        { type: 'tool_use', id: 'tu1', name: 'Bash', input: {} }
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:00:01Z' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: [
        { type: 'tool_result', tool_use_id: 'tu1', is_error: true, content: 'ENOENT: no such file or directory' }
      ] }, uuid: 'u3', timestamp: '2026-05-20T00:00:02Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    const errorEvents = ep.events.filter((e) => e.kind === 'error');
    expect(errorEvents).toHaveLength(1);
    expect(errorEvents[0].tool).toBe('Bash');
    expect(errorEvents[0].summary).toContain('ENOENT');
  });
  it('attributes error to "unknown" when tool_use_id not seen', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'делай' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: [
        { type: 'tool_result', tool_use_id: 'orphan', is_error: true, content: 'mystery error' }
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:00:02Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    const errs = ep.events.filter((e) => e.kind === 'error');
    expect(errs[0].tool).toBe('unknown');
    expect(errs[0].summary).toBe('mystery error');
  });
  it('truncates summary at 80 chars', () => {
    const longText = 'a'.repeat(200);
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'делай' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [
        { type: 'tool_use', id: 'tu1', name: 'Read', input: {} }
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:00:01Z' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: [
        { type: 'tool_result', tool_use_id: 'tu1', is_error: true, content: longText }
      ] }, uuid: 'u3', timestamp: '2026-05-20T00:00:02Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    const e = ep.events.find((x) => x.kind === 'error');
    expect(e.summary.length).toBe(80);
  });
  it('handles structured content array (text blocks)', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'делай' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [
        { type: 'tool_use', id: 'tu1', name: 'Edit', input: {} }
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:00:01Z' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: [
        { type: 'tool_result', tool_use_id: 'tu1', is_error: true, content: [{ type: 'text', text: 'String to replace not found' }] }
      ] }, uuid: 'u3', timestamp: '2026-05-20T00:00:02Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    const e = ep.events.find((x) => x.kind === 'error');
    expect(e.tool).toBe('Edit');
    expect(e.summary).toContain('String to replace');
  });
});

describe('promptText strips <system-reminder> blocks (Task 8)', () => {
  it('classifyTask is not polluted by reminder content', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user',
        content: '<system-reminder>почему как что зачем</system-reminder>\nрефактор биллинга'
      }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    expect(ep.primary_rationale.task_classification).toBe('refactor');
  });
  it('multiline system-reminder is stripped', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user',
        content: '<system-reminder>\nline 1\nline 2 with почему\n</system-reminder>\nfix баг'
      }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    expect(ep.primary_rationale.task_classification).toBe('bugfix');
  });
  it('multiple system-reminders all stripped', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user',
        content: '<system-reminder>почему</system-reminder>middle<system-reminder>как</system-reminder>создай фичу'
      }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    expect(ep.primary_rationale.task_classification).toBe('feature');
  });
  it('content array form also stripped', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user',
        content: [{ type: 'text', text: '<system-reminder>почему как</system-reminder> рефактор' }]
      }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    expect(ep.primary_rationale.task_classification).toBe('refactor');
  });
});

describe('classifyPromptSignal — extended dictionary (Task 9)', () => {
  it('detects "не совсем" as correction', () => {
    expect(classifyPromptSignal('не совсем то')).toBe('correction');
  });
  it('detects "wrong direction" as correction', () => {
    expect(classifyPromptSignal('wrong direction here')).toBe('correction');
  });
  it('detects "другое" as correction', () => {
    expect(classifyPromptSignal('другое решение нужно')).toBe('correction');
  });
  it('detects "класс" as approval', () => {
    expect(classifyPromptSignal('класс')).toBe('approval');
  });
  it('detects "well done" as approval', () => {
    expect(classifyPromptSignal('well done')).toBe('approval');
  });
  it('detects "nice" as approval', () => {
    expect(classifyPromptSignal('nice')).toBe('approval');
  });
  it('detects "теперь" prefix as new_task', () => {
    expect(classifyPromptSignal('теперь сделай биллинг')).toBe('new_task');
  });
  it('detects "далее" prefix as new_task', () => {
    expect(classifyPromptSignal('далее переходим к настройкам')).toBe('new_task');
  });
  it('detects "next" prefix as new_task', () => {
    expect(classifyPromptSignal('next step please')).toBe('new_task');
  });
  it('preserves existing — neutral for nondescript text', () => {
    expect(classifyPromptSignal('некий случайный текст')).toBe('neutral');
  });
  it('preserves existing — correction "переделай"', () => {
    expect(classifyPromptSignal('переделай это')).toBe('correction');
  });
});

import { parseReasoningTag } from './observer-transcript-parser.mjs';

describe('parseReasoningTag (Task 11)', () => {
  it('parses opt-in reasoning tag from assistant text', () => {
    const turn = [{ message: { role: 'assistant', content: [
      { type: 'text', text: '<!-- reasoning: triggers="Pravila §12.2;ADR-011" candidates="brain-retro;brainstorming" boundaries="Pravila §16" -->\nAnswer.' }
    ] } }];
    const tag = parseReasoningTag(turn);
    expect(tag).toEqual({
      triggers: ['Pravila §12.2', 'ADR-011'],
      candidates: ['brain-retro', 'brainstorming'],
      boundaries: ['Pravila §16'],
    });
  });
  it('returns null when no tag present', () => {
    expect(parseReasoningTag([{ message: { role: 'assistant', content: [{ type: 'text', text: 'plain' }] } }])).toBeNull();
  });
  it('safe on null/empty turn', () => {
    expect(parseReasoningTag(null)).toBeNull();
    expect(parseReasoningTag([])).toBeNull();
  });
  it('skips non-text blocks', () => {
    const turn = [{ message: { role: 'assistant', content: [{ type: 'tool_use', name: 'Read' }] } }];
    expect(parseReasoningTag(turn)).toBeNull();
  });
});

describe('parseTranscript — reasoning-tag merges with heuristic (Task 11)', () => {
  it('merges tag triggers into heuristic triggers (deduped)', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'делай' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [
        { type: 'text', text: 'Pravila §12.2 hard-rule\n<!-- reasoning: triggers="Pravila §12.2;ADR-011" candidates="brain-retro" boundaries="Pravila §16" -->' }
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:01:00Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    // heuristic finds 'Pravila §12.2' from text + tag adds 'ADR-011'; dedup
    expect(ep.primary_rationale.triggers_matched).toContain('Pravila §12.2');
    expect(ep.primary_rationale.triggers_matched).toContain('ADR-011');
    expect(ep.primary_rationale.triggers_matched.filter((t) => t === 'Pravila §12.2')).toHaveLength(1);
    // candidates: tag contributes 'brain-retro' (no numbered list in text)
    expect(ep.primary_rationale.candidates_considered).toContain('brain-retro');
    // boundaries: heuristic empty (no marker in plain text), tag adds 'Pravila §16'
    expect(ep.primary_rationale.boundaries_applied).toContain('Pravila §16');
  });
});

import { extractAgentInvocations } from './observer-transcript-parser.mjs';

describe('extractAgentInvocations (Task 12)', () => {
  it('emits subagent_invoked event from Agent tool_use', () => {
    const turn = [{ message: { role: 'assistant', content: [
      { type: 'tool_use', name: 'Agent',
        input: { subagent_type: 'general-purpose', model: 'sonnet', description: 'check files' } }
    ] } }];
    const ev = extractAgentInvocations(turn);
    expect(ev).toHaveLength(1);
    expect(ev[0].kind).toBe('subagent_invoked');
    expect(ev[0].subagent_type).toBe('general-purpose');
    expect(ev[0].model).toBe('sonnet');
    expect(ev[0].description).toBe('check files');
  });
  it('uses "unknown" when subagent_type missing', () => {
    const turn = [{ message: { role: 'assistant', content: [
      { type: 'tool_use', name: 'Agent', input: {} }
    ] } }];
    expect(extractAgentInvocations(turn)[0].subagent_type).toBe('unknown');
  });
  it('truncates description at 80 chars', () => {
    const turn = [{ message: { role: 'assistant', content: [
      { type: 'tool_use', name: 'Agent',
        input: { subagent_type: 'g', description: 'a'.repeat(200) } }
    ] } }];
    expect(extractAgentInvocations(turn)[0].description.length).toBe(80);
  });
  it('returns empty for non-Agent tool_use', () => {
    const turn = [{ message: { role: 'assistant', content: [
      { type: 'tool_use', name: 'Read', input: { file_path: '/a' } }
    ] } }];
    expect(extractAgentInvocations(turn)).toEqual([]);
  });
  it('safe on null/empty', () => {
    expect(extractAgentInvocations(null)).toEqual([]);
    expect(extractAgentInvocations([])).toEqual([]);
  });
});

describe('parseTranscript — subagent_invoked events (Task 12)', () => {
  it('emits subagent_invoked from Agent tool_use', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'делай' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [
        { type: 'tool_use', id: 't1', name: 'Agent',
          input: { subagent_type: 'tester', model: 'haiku', description: 'verify spec' } }
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:00:01Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    const subs = ep.events.filter((e) => e.kind === 'subagent_invoked');
    expect(subs).toHaveLength(1);
    expect(subs[0].subagent_type).toBe('tester');
    expect(subs[0].model).toBe('haiku');
  });
});

describe('parallel_session — pre-flight OR clause (Task 13 PIVOT)', () => {
  it('is true when Bash ran "git fetch && git log HEAD..origin/main"', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'pre-flight' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [
        { type: 'tool_use', id: 't1', name: 'Bash', input: { command: 'git fetch && git log HEAD..origin/main --oneline' } }
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:00:01Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    expect(ep.environment.parallel_session).toBe(true);
  });
  it('is true with origin/<other-branch> via HEAD..origin/feat', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'sync' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [
        { type: 'tool_use', id: 't1', name: 'Bash', input: { command: 'git fetch origin && git log HEAD..origin/feat/x --oneline' } }
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:00:01Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    expect(ep.environment.parallel_session).toBe(true);
  });
  it('F1 still works — collision text in tool_result triggers', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'go' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [
        { type: 'tool_use', id: 't1', name: 'Bash', input: { command: 'git commit -am ok' } }
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:00:01Z' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: [
        { type: 'tool_result', tool_use_id: 't1', content: 'fatal: index.lock exists' }
      ] }, uuid: 'u3', timestamp: '2026-05-20T00:00:02Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    expect(ep.environment.parallel_session).toBe(true);
  });
  it('false on regular Bash without pre-flight or collision', () => {
    const transcript = [
      JSON.stringify({ sessionId: 's' }),
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'go' }, uuid: 'u1', timestamp: '2026-05-20T00:00:00Z' }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [
        { type: 'tool_use', id: 't1', name: 'Bash', input: { command: 'npm run test:tools' } }
      ] }, uuid: 'u2', timestamp: '2026-05-20T00:00:01Z' }),
    ].join('\n');
    const ep = parseTranscript(transcript);
    expect(ep.environment.parallel_session).toBe(false);
  });
});

describe('parseTranscript v3 fields', () => {
  function transcriptWithSkill(skillName) {
    return [
      JSON.stringify({
        type: 'user',
        message: { role: 'user', content: 'добавь endpoint /api/foo' },
        timestamp: '2026-05-23T10:00:00Z',
        uuid: 'u-1',
        sessionId: 'sess-1',
      }),
      JSON.stringify({
        type: 'assistant',
        message: {
          role: 'assistant',
          content: [
            { type: 'tool_use', id: 't-1', name: 'Skill', input: { skill: skillName } },
          ],
        },
        timestamp: '2026-05-23T10:00:01Z',
        uuid: 'u-2',
        sessionId: 'sess-1',
      }),
    ].join('\n');
  }

  function transcriptDirectFeature() {
    return [
      JSON.stringify({
        type: 'user',
        message: { role: 'user', content: 'добавь новый endpoint /api/foo' },
        timestamp: '2026-05-23T10:00:00Z',
        uuid: 'u-1',
        sessionId: 'sess-1',
      }),
      JSON.stringify({
        type: 'assistant',
        message: { role: 'assistant', content: [{ type: 'text', text: 'делаю' }] },
        timestamp: '2026-05-23T10:00:01Z',
        uuid: 'u-2',
        sessionId: 'sess-1',
      }),
    ].join('\n');
  }

  function transcriptWithHookAttachment() {
    return [
      JSON.stringify({
        type: 'user',
        message: { role: 'user', content: 'ls' },
        timestamp: '2026-05-23T10:00:00Z',
        uuid: 'u-1',
        sessionId: 'sess-1',
      }),
      JSON.stringify({
        type: 'assistant',
        message: {
          role: 'assistant',
          content: [{ type: 'tool_use', id: 't-1', name: 'Bash', input: { command: 'ls' } }],
        },
        timestamp: '2026-05-23T10:00:01Z',
        uuid: 'u-2',
        sessionId: 'sess-1',
      }),
      JSON.stringify({
        type: 'attachment',
        attachment: { type: 'hook_success', hookName: 'PreToolUse:Bash', hookEvent: 'PreToolUse' },
        timestamp: '2026-05-23T10:00:01Z',
        uuid: 'u-3',
        sessionId: 'sess-1',
      }),
    ].join('\n');
  }

  it('emits schema_version: 4', () => {
    const ep = parseTranscript(transcriptDirectFeature(), 'sess-1');
    expect(ep.schema_version).toBe(4);
    expect(ep.schema_minor).toBe(4);
  });

  it('recommended_node is null on direct episode without router-state (no classifier signal)', () => {
    const ep = parseTranscript(transcriptDirectFeature(), 'sess-1');
    // Brain-retro #6 follow-up (2026-05-26): removed silent classification-map
    // fallback. recommended_node now reflects ONLY real classifier signal from
    // router-state. Direct feature episode without router-state → null, not '#19'.
    expect(ep.primary_rationale.recommended_node).toBeNull();
  });

  it('recommended_node is null when a skill was invoked', () => {
    const ep = parseTranscript(transcriptWithSkill('superpowers:writing-plans'), 'sess-1');
    expect(ep.primary_rationale.recommended_node).toBeNull();
  });

  it('hook_fired event includes both counts and scripts keys', () => {
    const ep = parseTranscript(transcriptWithHookAttachment(), 'sess-1');
    const hookEvent = ep.events.find((e) => e.kind === 'hook_fired');
    expect(hookEvent).toBeDefined();
    expect(hookEvent.counts).toBeDefined();
    expect(hookEvent.scripts).toBeDefined();
    expect(typeof hookEvent.scripts).toBe('object');
  });
});

describe('parseTranscript — router-state enrichment (Task 3)', () => {
  function makeTranscript(sessionId) {
    return [
      JSON.stringify({
        type: 'user',
        message: { role: 'user', content: 'добавь новый endpoint /api/bar' },
        timestamp: '2026-05-24T10:00:00Z',
        uuid: 'u-t3-1',
        sessionId,
      }),
      JSON.stringify({
        type: 'assistant',
        message: { role: 'assistant', content: [{ type: 'text', text: 'делаю' }] },
        timestamp: '2026-05-24T10:00:01Z',
        uuid: 'u-t3-2',
        sessionId,
      }),
    ].join('\n');
  }

  it('enriches primary_rationale from router-state file when present', () => {
    const dir = mkdtempSync(join(tmpdir(), 'router-state-test-'));
    const sessionId = 'test-session-t3-enrich';
    const state = {
      classification: { recommendedNode: '#42', recommendedChain: 'L13' },
      chainProgress: ['step-a', 'step-b'],
      chainCompleted: false,
    };
    writeFileSync(join(dir, `router-state-${sessionId}.json`), JSON.stringify(state));
    try {
      const ep = parseTranscript(makeTranscript(sessionId), sessionId, { routerStateBaseDir: dir });
      expect(ep.primary_rationale.recommended_node).toBe('#42');
      expect(ep.primary_rationale.recommended_chain).toBe('L13');
      expect(ep.primary_rationale.chain_progress).toEqual(['step-a', 'step-b']);
      expect(ep.primary_rationale.chain_completed).toBe(false);
    } finally {
      rmSync(dir, { recursive: true, force: true });
    }
  });

  it('recommended_node is null when router-state file is absent (no silent regex fallback)', () => {
    const dir = mkdtempSync(join(tmpdir(), 'router-state-test-'));
    const sessionId = 'test-session-t3-missing';
    try {
      const ep = parseTranscript(makeTranscript(sessionId), sessionId, { routerStateBaseDir: dir });
      // Brain-retro #6 follow-up (2026-05-26): no router-state → recommended_node = null.
      // The earlier silent classification-map fallback baked false-positive recommendations
      // into the JSONL — analysis showed 60-70% «recommended_node» values were just keyword
      // regex from classifyTask(prompt), not real classifier output.
      expect(ep.primary_rationale.recommended_node).toBeNull();
      expect(ep.primary_rationale.recommended_chain).toBeNull();
      expect(ep.primary_rationale.chain_progress).toEqual([]);
      expect(ep.primary_rationale.chain_completed).toBe(false);
    } finally {
      rmSync(dir, { recursive: true, force: true });
    }
  });

  it('merges router-state.task_cost (classifier tokens) into episode task_cost — A1 cost tracking', () => {
    const dir = mkdtempSync(join(tmpdir(), 'router-state-test-'));
    const sessionId = 'test-session-cost-merge';
    const state = {
      classification: { recommendedNode: '#19' },
      task_cost: {
        classifier_input_tokens: 4500,
        classifier_output_tokens: 120,
      },
    };
    writeFileSync(join(dir, `router-state-${sessionId}.json`), JSON.stringify(state));
    try {
      const ep = parseTranscript(makeTranscript(sessionId), sessionId, { routerStateBaseDir: dir });
      expect(ep.task_cost.classifier_input_tokens).toBe(4500);
      expect(ep.task_cost.classifier_output_tokens).toBe(120);
    } finally {
      rmSync(dir, { recursive: true, force: true });
    }
  });
});

// ─── Phase 3 deferred #2: parser write-block v4.3 ────────────────────────────

describe('parseTranscript — schema v4.3 write-block fields (phase 3 deferred #2)', () => {
  function simpleTranscript(prompt = 'add a feature', ts = '2026-05-25T10:00:00Z', sid = 's-v43') {
    return [
      JSON.stringify({ type: 'user', message: { role: 'user', content: prompt }, timestamp: ts, sessionId: sid }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [{ type: 'text', text: 'done' }] }, timestamp: ts, sessionId: sid }),
    ].join('\n');
  }

  it('emits schema_minor 4', () => {
    const ep = parseTranscript(simpleTranscript());
    expect(ep.schema_minor).toBe(4);
  });

  it('emits outcome_reviewed: null', () => {
    const ep = parseTranscript(simpleTranscript());
    expect('outcome_reviewed' in ep).toBe(true);
    expect(ep.outcome_reviewed).toBeNull();
  });

  it('emits outcome_reviewed_source: null', () => {
    const ep = parseTranscript(simpleTranscript());
    expect('outcome_reviewed_source' in ep).toBe(true);
    expect(ep.outcome_reviewed_source).toBeNull();
  });

  it('emits prompt_embedding_base64 as null when embedding model unavailable', () => {
    // parser is synchronous; embedding is null by design (filled async by stop-hook)
    const ep = parseTranscript(simpleTranscript());
    expect('prompt_embedding_base64' in ep).toBe(true);
    expect(ep.prompt_embedding_base64).toBeNull();
  });

  it('does not throw when transcript has unusual content', () => {
    // robustness guard: parser must never throw regardless of transcript shape
    expect(() => parseTranscript(simpleTranscript('', '2026-05-25T10:00:00Z'))).not.toThrow();
    expect(() => parseTranscript('')).not.toThrow();
    expect(() => parseTranscript('{ broken json\nnot valid')).not.toThrow();
  });

  it('task_cost has 8 new zero-default LLM-cost fields', () => {
    const ep = parseTranscript(simpleTranscript());
    const cost = ep.task_cost;
    expect(typeof cost.classifier_input_tokens).toBe('number');
    expect(typeof cost.classifier_output_tokens).toBe('number');
    expect(typeof cost.self_assessment_input_tokens).toBe('number');
    expect(typeof cost.self_assessment_output_tokens).toBe('number');
    expect(typeof cost.reviewer_input_tokens).toBe('number');
    expect(typeof cost.reviewer_output_tokens).toBe('number');
    expect(typeof cost.reviewer_subagent_usd).toBe('number');
    expect(typeof cost.reviewer_direct_fallback_usd).toBe('number');
    // all default to 0
    expect(cost.classifier_input_tokens).toBe(0);
    expect(cost.classifier_output_tokens).toBe(0);
    expect(cost.self_assessment_input_tokens).toBe(0);
    expect(cost.self_assessment_output_tokens).toBe(0);
    expect(cost.reviewer_input_tokens).toBe(0);
    expect(cost.reviewer_output_tokens).toBe(0);
    expect(cost.reviewer_subagent_usd).toBe(0);
    expect(cost.reviewer_direct_fallback_usd).toBe(0);
  });

  it('task_cost retains all existing fields alongside new ones', () => {
    const lines = [
      JSON.stringify({ type: 'user', message: { role: 'user', content: 'do it' } }),
      JSON.stringify({ type: 'assistant', message: { role: 'assistant', content: [{ type: 'text', text: 'ok' }], usage: { input_tokens: 100, output_tokens: 20, cache_read_input_tokens: 500, cache_creation_input_tokens: 50 } } }),
    ].join('\n');
    const cost = parseTranscript(lines).task_cost;
    expect(cost.input_tokens).toBe(100);
    expect(cost.output_tokens).toBe(20);
    expect(cost.cache_read_input_tokens).toBe(500);
    expect(cost.cache_creation_input_tokens).toBe(50);
    // new fields still 0 (populated retroactively by controller scripts)
    expect(cost.classifier_input_tokens).toBe(0);
    expect(cost.reviewer_subagent_usd).toBe(0);
  });
});

describe('classifyFilePath — Pass 3 path-pattern bucketing (project-brain-factor-analysis-4passes)', () => {
  it('classifies test files', () => {
    expect(classifyFilePath('tools/foo.test.mjs')).toBe('test');
    expect(classifyFilePath('app/tests/Feature/X.php')).toBe('test');
    expect(classifyFilePath('resources/js/foo.spec.ts')).toBe('test');
  });
  it('classifies config files', () => {
    expect(classifyFilePath('package.json')).toBe('config');
    expect(classifyFilePath('vite.config.ts')).toBe('config');
    expect(classifyFilePath('lefthook.yml')).toBe('config');
    expect(classifyFilePath('.env')).toBe('config');
    expect(classifyFilePath('tsconfig.json')).toBe('config');
  });
  it('classifies spec/plan files under docs/superpowers/', () => {
    expect(classifyFilePath('docs/superpowers/specs/x.md')).toBe('spec');
    expect(classifyFilePath('docs/superpowers/plans/x.md')).toBe('spec');
  });
  it('classifies normative documents', () => {
    expect(classifyFilePath('CLAUDE.md')).toBe('norm');
    expect(classifyFilePath('c:\\моя\\проекты\\портал crm\\Документация\\CLAUDE.md')).toBe('norm');
    expect(classifyFilePath('docs/Pravila_raboty_Claude_v1_1.md')).toBe('norm');
    expect(classifyFilePath('docs/Plugin_stack_rules_v1.md')).toBe('norm');
    expect(classifyFilePath('docs/Tooling_v8_3.md')).toBe('norm');
    expect(classifyFilePath('C:\\Users\\x\\.claude\\projects\\proj\\memory\\foo.md')).toBe('norm');
  });
  it('classifies data files', () => {
    expect(classifyFilePath('docs/observer/episodes-2026-05.jsonl')).toBe('data');
    expect(classifyFilePath('db/seed.csv')).toBe('data');
    expect(classifyFilePath('db/schema.sql')).toBe('data');
  });
  it('classifies app/tools source under src', () => {
    expect(classifyFilePath('app/Http/Controllers/X.php')).toBe('src');
    expect(classifyFilePath('tools/router-classifier.mjs')).toBe('src');
    expect(classifyFilePath('resources/js/views/Dashboard.vue')).toBe('src');
  });
  it('returns other for paths that fit no category', () => {
    expect(classifyFilePath('some-random-binary.png')).toBe('other');
  });
});

describe('classifyFilePath — config-driven normative stems (greenfield #3-observer)', () => {
  it('custom stem matches a greenfield normative doc', () => {
    expect(classifyFilePath('docs/MyRules_v2.md', ['MyRules'])).toBe('norm');
  });
  it('custom stems excluding Лидерра docs → those are not norm', () => {
    expect(classifyFilePath('docs/Pravila_raboty_Claude_v1_1.md', ['MyRules'])).toBe('other');
  });
  it('empty stems → Лидерра doc not norm, universal CLAUDE.md/memory still norm', () => {
    expect(classifyFilePath('docs/Tooling_v8_3.md', [])).toBe('other');
    expect(classifyFilePath('CLAUDE.md', [])).toBe('norm');
    expect(classifyFilePath('C:\\Users\\x\\.claude\\projects\\p\\memory\\f.md', [])).toBe('norm');
  });
  it('default (no stems arg) keeps current 3 Лидерра stems as norm', () => {
    expect(classifyFilePath('docs/Plugin_stack_rules_v1.md')).toBe('norm');
  });
  it('extractFileTypeDistribution threads custom normativeStems', () => {
    expect(extractFileTypeDistribution(['docs/MyRules_v2.md'], ['MyRules']).norm).toBe(1);
    expect(extractFileTypeDistribution(['docs/MyRules_v2.md'], []).norm).toBe(0);
  });
});

describe('extractFileTypeDistribution — Pass 3 (project-brain-factor-analysis-4passes)', () => {
  it('counts each path bucket and zero-fills missing categories', () => {
    const dist = extractFileTypeDistribution([
      'tools/router-classifier.mjs',
      'tools/router-classifier.test.mjs',
      'docs/superpowers/specs/x.md',
      'CLAUDE.md',
    ]);
    expect(dist.src).toBe(1);
    expect(dist.test).toBe(1);
    expect(dist.spec).toBe(1);
    expect(dist.norm).toBe(1);
    expect(dist.config).toBe(0);
    expect(dist.data).toBe(0);
    expect(dist.other).toBe(0);
  });
  it('returns all-zero distribution for empty input', () => {
    const dist = extractFileTypeDistribution([]);
    for (const v of Object.values(dist)) expect(v).toBe(0);
  });
});

describe('extractMcpServers — Pass 3 (project-brain-factor-analysis-4passes)', () => {
  it('extracts unique mcp__<server>__* prefixes from tool_use entries', () => {
    const turn = [
      assistantTurn([
        { type: 'tool_use', id: 't1', name: 'mcp__github__list_issues', input: {} },
        { type: 'tool_use', id: 't2', name: 'mcp__github__get_pr', input: {} },
        { type: 'tool_use', id: 't3', name: 'mcp__playwright__browser_click', input: {} },
        { type: 'tool_use', id: 't4', name: 'Read', input: { file_path: 'a.txt' } },
      ], '2026-05-25T10:00:00Z'),
    ];
    expect(extractMcpServers(turn).sort()).toEqual(['github', 'playwright']);
  });
  it('returns empty array when no mcp tools used', () => {
    const turn = [assistantTurn([{ type: 'tool_use', id: 't1', name: 'Read', input: { file_path: 'a' } }], '2026-05-25T10:00:00Z')];
    expect(extractMcpServers(turn)).toEqual([]);
  });
});

describe('parseTranscript — Pass 3 task_meta block (project-brain-factor-analysis-4passes)', () => {
  it('includes prompt_length_chars / mcp_servers_used / file_type_distribution under task_meta', () => {
    const t = jsonl([
      userPrompt('добавь функцию X в файл tools/router-classifier.mjs', '2026-05-25T10:00:00Z'),
      assistantTurn([
        { type: 'tool_use', id: 't1', name: 'mcp__github__get_pr', input: {} },
        { type: 'tool_use', id: 't2', name: 'Read', input: { file_path: 'tools/router-classifier.mjs' } },
        { type: 'tool_use', id: 't3', name: 'Edit', input: { file_path: 'tools/router-classifier.test.mjs', old_string: 'a', new_string: 'b' } },
      ], '2026-05-25T10:01:00Z'),
    ]);
    const ep = parseTranscript(t);
    expect(ep.task_meta).toBeDefined();
    expect(ep.task_meta.prompt_length_chars).toBe('добавь функцию X в файл tools/router-classifier.mjs'.length);
    expect(ep.task_meta.mcp_servers_used).toEqual(['github']);
    expect(ep.task_meta.file_type_distribution.src).toBe(1);
    expect(ep.task_meta.file_type_distribution.test).toBe(1);
  });

  it('task_meta is present even on empty transcript (null-safe defaults)', () => {
    const ep = parseTranscript('');
    expect(ep.task_meta).toBeDefined();
    expect(ep.task_meta.prompt_length_chars).toBe(0);
    expect(ep.task_meta.mcp_servers_used).toEqual([]);
    expect(ep.task_meta.file_type_distribution.other).toBe(0);
  });
});