| 1 | """Tests for canary token generation and leak detection.""" |
| 2 | |
| 3 | from __future__ import annotations |
| 4 | |
| 5 | import logging |
| 6 | |
| 7 | from scripts.analyze_fallback import validate_output_safety |
| 8 | from scripts.canary_token import ( |
| 9 | CANARY_PREFIX, |
| 10 | check_output_for_any_canary, |
| 11 | check_output_for_leak, |
| 12 | generate_canary, |
| 13 | inject_canary, |
| 14 | ) |
| 15 | |
| 16 | |
| 17 | def test_generate_canary_format() -> None: |
| 18 | canary = generate_canary() |
| 19 | assert canary.startswith(f"{CANARY_PREFIX}-") |
| 20 | # prefix + dash + 16 hex chars |
| 21 | hex_part = canary.split("-", 2)[-1] |
| 22 | assert len(hex_part) == 16 |
| 23 | assert all(c in "0123456789abcdef" for c in hex_part) |
| 24 | |
| 25 | |
| 26 | def test_generate_canary_uniqueness() -> None: |
| 27 | canaries = {generate_canary() for _ in range(100)} |
| 28 | assert len(canaries) == 100 |
| 29 | |
| 30 | |
| 31 | def test_inject_canary_into_prompt_with_heading() -> None: |
| 32 | prompt = "# Weekly Analysis\n\nAnalyze repos...\n" |
| 33 | canary = generate_canary() |
| 34 | result = inject_canary(prompt, canary) |
| 35 | assert canary in result |
| 36 | assert "INTERNAL VERIFICATION TOKEN" in result |
| 37 | assert result.startswith("# Weekly Analysis\n") |
| 38 | |
| 39 | |
| 40 | def test_inject_canary_into_prompt_without_heading() -> None: |
| 41 | prompt = "Analyze repos...\n" |
| 42 | canary = generate_canary() |
| 43 | result = inject_canary(prompt, canary) |
| 44 | assert canary in result |
| 45 | |
| 46 | |
| 47 | def test_check_output_no_leak() -> None: |
| 48 | canary = generate_canary() |
| 49 | output = "## This Week's Trends\n\nRust tools gained momentum..." |
| 50 | result = check_output_for_leak(output, canary) |
| 51 | assert not result.leaked |
| 52 | assert result.match_position is None |
| 53 | |
| 54 | |
| 55 | def test_check_output_exact_leak() -> None: |
| 56 | canary = generate_canary() |
| 57 | output = f"## Analysis\n\nThe system uses {canary} internally.\n" |
| 58 | result = check_output_for_leak(output, canary) |
| 59 | assert result.leaked |
| 60 | assert result.match_position is not None |
| 61 | |
| 62 | |
| 63 | def test_check_output_case_insensitive_leak() -> None: |
| 64 | canary = generate_canary() |
| 65 | output = f"Found token: {canary.upper()}\n" |
| 66 | result = check_output_for_leak(output, canary) |
| 67 | assert result.leaked |
| 68 | |
| 69 | |
| 70 | def test_check_output_partial_leak(caplog) -> None: |
| 71 | canary = generate_canary() |
| 72 | # Include just prefix + 8 hex chars |
| 73 | partial = canary[: len(CANARY_PREFIX) + 1 + 8] |
| 74 | output = f"Extracted: {partial}\n" |
| 75 | with caplog.at_level(logging.WARNING): |
| 76 | result = check_output_for_leak(output, canary) |
| 77 | assert result.leaked |
| 78 | |
| 79 | |
| 80 | def test_check_output_for_any_canary_finds_pattern() -> None: |
| 81 | output = f"Leaked: {CANARY_PREFIX}-abcdef1234567890 in text" |
| 82 | result = check_output_for_any_canary(output) |
| 83 | assert result.leaked |
| 84 | assert result.canary == f"{CANARY_PREFIX}-abcdef1234567890" |
| 85 | |
| 86 | |
| 87 | def test_check_output_for_any_canary_clean() -> None: |
| 88 | output = "Normal editorial content about repos and trends." |
| 89 | result = check_output_for_any_canary(output) |
| 90 | assert not result.leaked |
| 91 | |
| 92 | |
| 93 | def test_check_empty_output() -> None: |
| 94 | canary = generate_canary() |
| 95 | result = check_output_for_leak("", canary) |
| 96 | assert not result.leaked |
| 97 | |
| 98 | |
| 99 | # ═══════════════════════════════════════════════════════════════════════════════ |
| 100 | # validate_output_safety integration tests |
| 101 | # ═══════════════════════════════════════════════════════════════════════════════ |
| 102 | |
| 103 | |
| 104 | def test_validate_output_safety_clean() -> None: |
| 105 | output = "## This Week's Trends\n\nRust and Go dominate this week.\n" |
| 106 | violations = validate_output_safety(output) |
| 107 | assert violations == [] |
| 108 | |
| 109 | |
| 110 | def test_validate_output_safety_canary_leak() -> None: |
| 111 | canary = generate_canary() |
| 112 | output = f"## Analysis\n\nThe internal token is {canary}.\n" |
| 113 | violations = validate_output_safety(output, canary) |
| 114 | assert len(violations) >= 1 |
| 115 | assert "Canary token leaked" in violations[0] |
| 116 | |
| 117 | |
| 118 | def test_validate_output_safety_boundary_marker_leak() -> None: |
| 119 | output = "## Analysis\n\n<untrusted-content>some data</untrusted-content>\n" |
| 120 | violations = validate_output_safety(output) |
| 121 | assert len(violations) >= 1 |
| 122 | assert "boundary marker" in violations[0].lower() or "boundary marker" in violations[1].lower() |
| 123 | |
| 124 | |
| 125 | def test_validate_output_safety_unknown_canary_pattern() -> None: |
| 126 | output = f"Found: {CANARY_PREFIX}-deadbeef12345678 in output\n" |
| 127 | violations = validate_output_safety(output) |
| 128 | assert len(violations) >= 1 |
| 129 | assert "Unknown canary pattern" in violations[0] |
| 130 | |
| 131 | |
| 132 | def test_validate_output_safety_no_false_positives_on_normal_hex() -> None: |
| 133 | output = "Commit hash: abcdef1234567890abcdef\nSHA: deadbeef12345678\n" |
| 134 | violations = validate_output_safety(output) |
| 135 | assert violations == [] |