repopilot 0.13.0

Local-first CLI for repository audit, architecture risk detection, baseline tracking, and CI-friendly code review.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72

# Report Schema Migration

RepoPilot report schemas are intentionally versioned because JSON reports are
useful in CI, dashboards, PR bots, and downstream tooling.

## Current direction

Schema `0.17` is the current scan report schema for RepoPilot 0.13.0. It keeps
`0.15` provenance and `risk-v3` fields, keeps `0.16` context graph diagnostics,
and adds raw-vs-visible finding and signal-quality metrics.

| Field | Where | Why |
|---|---|---|
| `raw_findings_count` | scan, baseline-scan | Shows how many findings existed before default visibility filtering. |
| `visible_findings_count` | scan, baseline-scan | Shows the findings rendered in the current report. |
| `hidden_suggestions_count` | scan, baseline-scan | Shows strict-only suggestions hidden from the default report. |
| `raw_signal_quality` | scan, baseline-scan, review | Summarizes quality before visibility filtering. |
| `visible_signal_quality` | scan, baseline-scan, review | Summarizes quality for rendered findings. |
| `signal_quality` | scan, baseline-scan, review | Compatibility alias for `visible_signal_quality`. |
| `provenance` | finding | Explains detector, lifecycle, signal source, and analysis scope. |
| `risk.signals[].source` | finding risk signal | Gives machine-readable source families for risk explanations. |
| `scan_timings.contract_validation_us` | scan timings | Exposes finding-contract validation timing separately. |

RepoPilot's current reader accepts `0.15`, `0.16`, and `0.17` scan reports
during the transition. New reports are emitted as `0.17`.

Schema `0.14` added optional local feedback transparency:

| Field | Where | Why |
|---|---|---|
| `local_feedback` | scan, baseline-scan, review, receipt | Shows how many `.repopilot/feedback.yml` suppressions were loaded, matched, unmatched, invalid, or blocked by parse errors. |

The field is omitted when no local feedback file was applied or when the command
uses `--ignore-feedback`.

## 0.13 accounting migration

Schema `0.13` clarifies scan accounting names:

| Old field | New field | Why |
|---|---|---|
| `files_count` | `files_analyzed` | The value counts analyzed text files, not all discovered files. |
| `lines_of_code` | `non_empty_lines` | RepoPilot counts non-empty lines used by audit thresholds. |
| `skipped_files_count` | `large_files_skipped` | The field describes large files skipped by size/limit rules. |

## Consumer guidance

Consumers should prefer:

```json
{
  "files_analyzed": 42,
  "non_empty_lines": 3200,
  "large_files_skipped": 1
}
```

If a downstream consumer still needs historical reports, keep that compatibility
in the consumer instead of relying on RepoPilot's current reader:

```text
files_analyzed <- files_count
non_empty_lines <- lines_of_code
large_files_skipped <- skipped_files_count
```

## Compatibility policy

RepoPilot `0.13.0` intentionally tightens current report readers. Commands such
as `repopilot compare` require current scan reports with both top-level
`schema_version` and `report.schema_version` matching the current schema.
Baseline files keep their separate baseline schema policy.